Open MHA Developer Manual

User Manual:

Open the PDF directly: View PDF .
Page Count: 1221 [warning: Documents this large are best viewed by clicking the View PDF Link!]

Overview
- Structure
- Platform Services and Conventions
Todo List
Module Documentation
Namespace Documentation
Class Documentation
File Documentation
Index

The Open Master Hearing Aid

(openMHA)

4.8.0

Plugin Developers’ Manual

The Open Master Hearing Aid (openMHA) – Plugin Developers’ Manual

HörTech gGmbH

Marie-Curie-Str. 2

D–26129 Oldenburg

iii

LICENSE AGREEMENT

This ﬁle is part of the HörTech Open Master Hearing Aid (openMHA)

openMHA is free software: you can redistribute it and/or modify it under the terms of the GNU

Affero General Public License as published by the Free Software Foundation, version 3 of the

License.

openMHA is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;

without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR

PURPOSE. See the GNU Affero General Public License, version 3 for more details.

You should have received a copy of the GNU Affero General Public License, version 3 along

with openMHA. If not, see <http://www.gnu.org/licenses/>.

Contents

1 Overview 1

1.1 Structure ........................................ 1

1.2 Platform Services and Conventions ......................... 2

2 Todo List 4

3 Module Documentation 4

3.1 Concept of Variables and Data Exchange in the openMHA ............ 4

3.2 The openMHA Plugins (programming interface) .................. 7

3.3 Writing openMHA Plugins. A step-by-step tutorial ................. 11

3.4 The MHA Framework interface ............................ 27

3.5 Communication between algorithms ......................... 28

3.6 Error handling in the openMHA ........................... 32

3.7 The openMHA conﬁguration language ....................... 34

3.8 The openMHA Toolbox library ............................ 35

3.9 Vector and matrix processing toolbox ........................ 37

3.10 Complex arithmetics in the openMHA ........................ 61

3.11 Fast Fourier Transform functions ........................... 71

4 Namespace Documentation 78

4.1 ac2lsl Namespace Reference ............................ 78

4.2 acmon Namespace Reference ............................ 78

4.3 acsave Namespace Reference ............................ 79

4.4 addsndﬁle Namespace Reference .......................... 79

4.5 ADM Namespace Reference ............................. 80

4.6 AuditoryProﬁle Namespace Reference ....................... 81

4.7 coherence Namespace Reference .......................... 82

4.8 dc Namespace Reference .............................. 82

4.9 dc_simple Namespace Reference .......................... 83

4.10 delay Namespace Reference ............................. 84

4.11 delaysum Namespace Reference .......................... 84

4.12 DynComp Namespace Reference .......................... 85

4.13 fader_wave Namespace Reference ......................... 86

4.14 fftfbpow Namespace Reference ........................... 86

4.15 fftﬁlterbank Namespace Reference ......................... 87

4.16 fshift Namespace Reference ............................. 87

4.17 fshift_hilbert Namespace Reference ......................... 88

4.18 gain Namespace Reference ............................. 88

4.19 gtfb_analyzer Namespace Reference ........................ 88

4.20 matrixmixer Namespace Reference ......................... 89

4.21 MHA_AC Namespace Reference .......................... 89

4.22 mha_error_helpers Namespace Reference ..................... 90

4.23 MHA_TCP Namespace Reference ......................... 91

4.24 mhachain Namespace Reference .......................... 93

4.25 MHAEvents Namespace Reference ......................... 93

4.26 MHAFilter Namespace Reference .......................... 94

4.27 MHAIOJack Namespace Reference ......................... 98

4.28 MHAIOJackdb Namespace Reference ....................... 98

4.29 MHAIOPortAudio Namespace Reference ...................... 98

4.30 MHAJack Namespace Reference .......................... 99

CONTENTS v

4.31 MHAKernel Namespace Reference .........................101

4.32 MHAMultiSrc Namespace Reference ........................101

4.33 MHAOvlFilter Namespace Reference ........................102

4.34 MHAOvlFilter::barkscale Namespace Reference ..................103

4.35 MHAOvlFilter::FreqScaleFun Namespace Reference ...............103

4.36 MHAOvlFilter::ShapeFun Namespace Reference .................106

4.37 MHAParser Namespace Reference .........................107

4.38 MHAParser::StrCnv Namespace Reference ....................112

4.39 MHAPlugin Namespace Reference .........................116

4.40 MHAPlugin_Resampling Namespace Reference ..................117

4.41 MHAPlugin_Split Namespace Reference ......................117

4.42 MHASignal Namespace Reference .........................118

4.43 MHASndFile Namespace Reference ........................125

4.44 MHATableLookup Namespace Reference .....................125

4.45 MHAWindow Namespace Reference ........................125

4.46 multibandcompressor Namespace Reference ...................127

4.47 noisePowProposedScale Namespace Reference .................127

4.48 overlapadd Namespace Reference .........................127

4.49 plingploing Namespace Reference .........................128

4.50 PluginLoader Namespace Reference ........................128

4.51 route Namespace Reference .............................129

4.52 shadowﬁlter_begin Namespace Reference .....................129

4.53 shadowﬁlter_end Namespace Reference ......................129

4.54 smoothgains_bridge Namespace Reference ....................130

4.55 testplugin Namespace Reference ..........................130

5 Class Documentation 130

5.1 ac2lsl::ac2lsl_t Class Reference ...........................130

5.2 ac2lsl::cfg_t Class Reference ............................133

5.3 ac2lsl::save_var_base_t Class Reference ......................136

5.4 ac2lsl::save_var_t<T>Class Template Reference ................137

5.5 ac2lsl::save_var_t<mha_complex_t >Class Template Reference ........140

5.6 ac2lsl::type_info Struct Reference ..........................143

5.7 ac2osc_t Class Reference ..............................144

5.8 ac2wave_if_t Class Reference ............................148

5.9 ac2wave_t Class Reference .............................150

5.10 acConcat_wave Class Reference ..........................151

5.11 acConcat_wave_conﬁg Class Reference ......................154

5.12 acmon::ac_monitor_t Class Reference .......................155

5.13 acmon::acmon_t Class Reference ..........................157

5.14 acPooling_wave Class Reference ..........................160

5.15 acPooling_wave_conﬁg Class Reference ......................163

5.16 acsave::acsave_t Class Reference .........................165

5.17 acsave::cfg_t Class Reference ............................168

5.18 acsave::mat4head_t Struct Reference ........................169

5.19 acsave::save_var_t Class Reference ........................170

5.20 acSteer Class Reference ...............................172

5.21 acSteer_conﬁg Class Reference ...........................174

5.22 acTransform_wave Class Reference .........................176

5.23 acTransform_wave_conﬁg Class Reference ....................178

5.24 addsndﬁle::addsndﬁle_if_t Class Reference ....................180

5.25 addsndﬁle::level_adapt_t Class Reference .....................183

vi CONTENTS

5.26 addsndﬁle::resampled_soundﬁle_t Class Reference ................184

5.27 addsndﬁle::sndﬁle_t Class Reference ........................186

5.28 addsndﬁle::waveform_proxy_t Class Reference ..................187

5.29 ADM::ADM<F>Class Template Reference ....................188

5.30 ADM::Delay<F>Class Template Reference ...................191

5.31 ADM::Linearphase_FIR<F>Class Template Reference .............193

5.32 adm_if_t Class Reference ..............................195

5.33 adm_rtconﬁg_t Class Reference ...........................197

5.34 algo_comm_t Struct Reference ...........................201

5.35 alsa_base_t Class Reference ............................206

5.36 alsa_dev_par_parser_t Class Reference ......................208

5.37 alsa_t<T>Class Template Reference .......................210

5.38 altplugs_t Class Reference ..............................213

5.39 analysepath_t Class Reference ...........................217

5.40 analysispath_if_t Class Reference ..........................219

5.41 AuditoryProﬁle::fmap_t Class Reference ......................221

5.42 AuditoryProﬁle::parser_t Class Reference .....................222

5.43 AuditoryProﬁle::parser_t::ear_t Class Reference ..................223

5.44 AuditoryProﬁle::parser_t::fmap_t Class Reference .................224

5.45 AuditoryProﬁle::proﬁle_t Class Reference ......................226

5.46 AuditoryProﬁle::proﬁle_t::ear_t Class Reference ..................227

5.47 bbcalib_interface_t Class Reference .........................228

5.48 calibrator_runtime_layer_t Class Reference ....................229

5.49 calibrator_t Class Reference .............................231

5.50 calibrator_variables_t Class Reference .......................233

5.51 cfg_t Class Reference ................................235

5.52 coherence::cohﬂt_if_t Class Reference .......................237

5.53 coherence::cohﬂt_t Class Reference ........................239

5.54 coherence::vars_t Class Reference .........................241

5.55 combc_if_t Class Reference .............................243

5.56 combc_t Class Reference ..............................244

5.57 comm_var_t Struct Reference ............................246

5.58 cpuload_t Class Reference ..............................247

5.59 db_if_t Class Reference ...............................249

5.60 db_t Class Reference .................................251

5.61 dc::dc_if_t Class Reference .............................253

5.62 dc::dc_t Class Reference ...............................255

5.63 dc::dc_vars_t Class Reference ............................257

5.64 dc::dc_vars_validator_t Class Reference ......................260

5.65 dc::wb_inhib_cfg_t Class Reference .........................260

5.66 dc::wideband_inhib_vars_t Class Reference ....................261

5.67 dc_simple::dc_if_t Class Reference .........................263

5.68 dc_simple::dc_t Class Reference ..........................266

5.69 dc_simple::dc_t::line_t Class Reference .......................268

5.70 dc_simple::dc_vars_t Class Reference .......................269

5.71 dc_simple::dc_vars_validator_t Class Reference ..................270

5.72 dc_simple::level_smoother_t Class Reference ...................271

5.73 delay::interface_t Class Reference ..........................273

5.74 delaysum::delaysum_if_t Class Reference .....................274

5.75 delaysum::delaysum_t Class Reference .......................277

5.76 doasvm_classiﬁcation Class Reference .......................279

CONTENTS vii

5.77 doasvm_classiﬁcation_conﬁg Class Reference ...................281

5.78 doasvm_feature_extraction Class Reference ....................283

5.79 doasvm_feature_extraction_conﬁg Class Reference ................285

5.80 droptect_t Class Reference ..............................287

5.81 ds_t Class Reference .................................290

5.82 dynamiclib_t Class Reference ............................291

5.83 DynComp::dc_afterburn_rt_t Class Reference ...................292

5.84 DynComp::dc_afterburn_t Class Reference .....................294

5.85 DynComp::dc_afterburn_vars_t Class Reference .................296

5.86 DynComp::gaintable_t Class Reference .......................298

5.87 example1_t Class Reference .............................303

5.88 example2_t Class Reference .............................305

5.89 example3_t Class Reference .............................308

5.90 example4_t Class Reference .............................312

5.91 example5_t Class Reference .............................315

5.92 example6_t Class Reference .............................316

5.93 expression_t Class Reference ............................318

5.94 fader_if_t Class Reference ..............................318

5.95 fader_wave::fader_wave_if_t Class Reference ...................320

5.96 fader_wave::level_adapt_t Class Reference ....................322

5.97 fftfbpow::fftfbpow_interface_t Class Reference ...................323

5.98 fftfbpow::fftfbpow_t Class Reference ........................326

5.99 fftﬁlterbank::fftfb_interface_t Class Reference ...................328

5.100 fftﬁlterbank::fftfb_plug_t Class Reference ......................330

5.101 fshift::fshift_conﬁg_t Class Reference ........................332

5.102 fshift::fshift_t Class Reference ............................334

5.103 fshift_hilbert::frequency_translator_t Class Reference ...............337

5.104 fshift_hilbert::hilbert_shifter_t Class Reference ...................339

5.105 fw_t Class Reference .................................343

5.106 fw_vars_t Class Reference ..............................348

5.107 gain::gain_if_t Class Reference ...........................350

5.108 gain::scaler_t Class Reference ............................352

5.109 gtfb_analyzer::gtfb_analyzer_cfg_t Struct Reference ...............352

5.110 gtfb_analyzer::gtfb_analyzer_t Class Reference ..................355

5.111 hanning_ramps_t Class Reference .........................358

5.112 identity_t Class Reference ..............................359

5.113 iirﬁlter_t Class Reference ...............................360

5.114 io_alsa_t Class Reference ..............................361

5.115 io_ﬁle_t Class Reference ...............................365

5.116 io_lib_t Class Reference ...............................369

5.117 io_parser_t Class Reference .............................372

5.118 io_tcp_fwcb_t Class Reference ...........................375

5.119 io_tcp_parser_t Class Reference ..........................378

5.120 io_tcp_sound_t Class Reference ...........................384

5.121 io_tcp_sound_t::ﬂoat_union Union Reference ...................388

5.122 io_tcp_t Class Reference ...............................389

5.123 latex_doc_t Class Reference .............................391

5.124 lpc Class Reference ..................................395

5.125 lpc_bl_predictor Class Reference ..........................398

5.126 lpc_bl_predictor_conﬁg Class Reference ......................400

5.127 lpc_burglattice Class Reference ...........................402

viii CONTENTS

5.128 lpc_burglattice_conﬁg Class Reference .......................405

5.129 lpc_conﬁg Class Reference .............................406

5.130 matrixmixer::cfg_t Class Reference .........................408

5.131 matrixmixer::matmix_t Class Reference .......................410

5.132 MHA_AC::ac2matrix_helper_t Class Reference ..................412

5.133 MHA_AC::ac2matrix_t Class Reference .......................413

5.134 MHA_AC::acspace2matrix_t Class Reference ...................415

5.135 MHA_AC::double_t Class Reference ........................418

5.136 MHA_AC::ﬂoat_t Class Reference ..........................420

5.137 MHA_AC::int_t Class Reference ...........................421

5.138 MHA_AC::spectrum_t Class Reference .......................422

5.139 MHA_AC::stat_t Class Reference ..........................424

5.140 MHA_AC::waveform_t Class Reference .......................425

5.141 mha_audio_descriptor_t Struct Reference .....................427

5.142 mha_audio_t Struct Reference ............................428

5.143 mha_channel_info_t Struct Reference ........................429

5.144 mha_complex_t Struct Reference ..........................431

5.145 mha_complex_test_array_t Struct Reference ....................431

5.146 mha_dblbuf_t<FIFO >Class Template Reference ................432

5.147 mha_direction_t Struct Reference ..........................438

5.148 mha_drifter_ﬁfo_t<T>Class Template Reference ................439

5.149 MHA_Error Class Reference .............................445

5.150 mha_ﬁfo_lw_t<T>Class Template Reference ..................447

5.151 mha_ﬁfo_posix_threads_t Class Reference .....................450

5.152 mha_ﬁfo_t<T>Class Template Reference ....................452

5.153 mha_ﬁfo_thread_guard_t Class Reference .....................456

5.154 mha_ﬁfo_thread_platform_t Class Reference ....................457

5.155 mha_real_test_array_t Struct Reference ......................460

5.156 mha_rt_ﬁfo_element_t<T>Class Template Reference .............460

5.157 mha_rt_ﬁfo_t<T>Class Template Reference ...................462

5.158 mha_spec_t Struct Reference ............................465

5.159 mha_stash_environment_variable_t Class Reference ...............466

5.160 MHA_TCP::Async_Notify Class Reference .....................468

5.161 MHA_TCP::Client Class Reference .........................469

5.162 MHA_TCP::Connection Class Reference ......................471

5.163 MHA_TCP::Event_Watcher Class Reference ....................478

5.164 MHA_TCP::OS_EVENT_TYPE Struct Reference .................480

5.165 MHA_TCP::Server Class Reference .........................481

5.166 MHA_TCP::sock_init_t Class Reference ......................483

5.167 MHA_TCP::Sockaccept_Event Class Reference ..................484

5.168 MHA_TCP::Sockread_Event Class Reference ...................485

5.169 MHA_TCP::Sockwrite_Event Class Reference ...................486

5.170 MHA_TCP::Thread Class Reference ........................486

5.171 MHA_TCP::Timeout_Event Class Reference ....................490

5.172 MHA_TCP::Timeout_Watcher Class Reference ..................491

5.173 MHA_TCP::Wakeup_Event Class Reference ....................493

5.174 mha_tictoc_t Struct Reference ............................495

5.175 mha_wave_t Struct Reference ............................496

5.176 mhachain::chain_base_t Class Reference .....................497

5.177 mhachain::mhachain_t Class Reference ......................500

5.178 mhachain::plugs_t Class Reference .........................501

CONTENTS ix

5.179 mhaconﬁg_t Struct Reference ............................504

5.180 MHAEvents::connector_base_t Class Reference ..................506

5.181 MHAEvents::connector_t<receiver_t >Class Template Reference .......508

5.182 MHAEvents::emitter_t Class Reference .......................510

5.183 MHAEvents::patchbay_t<receiver_t >Class Template Reference ........512

5.184 MHAFilter::adapt_ﬁlter_param_t Class Reference .................513

5.185 MHAFilter::adapt_ﬁlter_state_t Class Reference ..................514

5.186 MHAFilter::adapt_ﬁlter_t Class Reference .....................515

5.187 MHAFilter::blockprocessing_polyphase_resampling_t Class Reference . . . . . 517

5.188 MHAFilter::complex_bandpass_t Class Reference .................520

5.189 MHAFilter::diff_t Class Reference ..........................522

5.190 MHAFilter::fftﬁlter_t Class Reference ........................523

5.191 MHAFilter::fftﬁlterbank_t Class Reference .....................527

5.192 MHAFilter::ﬁlter_t Class Reference .........................531

5.193 MHAFilter::gamma_ﬂt_t Class Reference ......................535

5.194 MHAFilter::iir_ﬁlter_state_t Class Reference ....................538

5.195 MHAFilter::iir_ﬁlter_t Class Reference ........................539

5.196 MHAFilter::iir_ord1_real_t Class Reference .....................542

5.197 MHAFilter::o1_ar_ﬁlter_t Class Reference .....................545

5.198 MHAFilter::o1ﬂt_lowpass_t Class Reference ....................548

5.199 MHAFilter::o1ﬂt_maxtrack_t Class Reference ...................551

5.200 MHAFilter::o1ﬂt_mintrack_t Class Reference ....................552

5.201 MHAFilter::partitioned_convolution_t Class Reference ...............554

5.202 MHAFilter::partitioned_convolution_t::index_t Struct Reference ..........558

5.203 MHAFilter::polyphase_resampling_t Class Reference ...............559

5.204 MHAFilter::resampling_ﬁlter_t Class Reference ..................564

5.205 MHAFilter::smoothspec_t Class Reference .....................565

5.206 MHAFilter::thirdoctave_analyzer_t Class Reference ................568

5.207 MHAFilter::transfer_function_t Struct Reference ..................570

5.208 MHAFilter::transfer_matrix_t Struct Reference ...................573

5.209 MHAIOJack::io_jack_t Class Reference .......................574

5.210 MHAIOJackdb::io_jack_t Class Reference .....................578

5.211 MHAIOPortAudio::device_info_t Class Reference .................583

5.212 MHAIOPortAudio::io_portaudio_t Class Reference .................585

5.213 MHAJack::client_avg_t Class Reference ......................588

5.214 MHAJack::client_noncont_t Class Reference ....................591

5.215 MHAJack::client_t Class Reference .........................593

5.216 MHAJack::port_t Class Reference ..........................599

5.217 MHAKernel::algo_comm_class_t Class Reference .................603

5.218 MHAKernel::comm_var_map_t Class Reference ..................606

5.219 MHAMultiSrc::base_t Class Reference .......................607

5.220 MHAMultiSrc::channel_t Class Reference .....................608

5.221 MHAMultiSrc::channels_t Class Reference .....................609

5.222 MHAMultiSrc::spectrum_t Class Reference .....................610

5.223 MHAMultiSrc::waveform_t Class Reference ....................611

5.224 MHAOvlFilter::band_descriptor_t Class Reference .................612

5.225 MHAOvlFilter::barkscale::bark2hz_t Class Reference ...............613

5.226 MHAOvlFilter::barkscale::hz2bark_t Class Reference ...............614

5.227 MHAOvlFilter::fftfb_ac_info_t Class Reference ...................615

5.228 MHAOvlFilter::fftfb_t Class Reference ........................616

5.229 MHAOvlFilter::fftfb_vars_t Class Reference .....................619

x CONTENTS

5.230 MHAOvlFilter::fscale_bw_t Class Reference ....................622

5.231 MHAOvlFilter::fscale_t Class Reference ......................623

5.232 MHAOvlFilter::fspacing_t Class Reference .....................625

5.233 MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t Class Reference .......627

5.234 MHAOvlFilter::overlap_save_ﬁlterbank_t Class Reference ............628

5.235 MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t Class Reference ........630

5.236 MHAOvlFilter::scale_var_t Class Reference ....................632

5.237 MHAParser::base_t Class Reference ........................634

5.238 MHAParser::base_t::replace_t Class Reference ..................643

5.239 MHAParser::bool_mon_t Class Reference .....................644

5.240 MHAParser::bool_t Class Reference ........................645

5.241 MHAParser::c_ifc_parser_t Class Reference ....................648

5.242 MHAParser::commit_t<receiver_t >Class Template Reference .........650

5.243 MHAParser::complex_mon_t Class Reference ...................652

5.244 MHAParser::complex_t Class Reference ......................654

5.245 MHAParser::entry_t Class Reference ........................656

5.246 MHAParser::expression_t Class Reference .....................657

5.247 MHAParser::ﬂoat_mon_t Class Reference .....................658

5.248 MHAParser::ﬂoat_t Class Reference ........................659

5.249 MHAParser::int_mon_t Class Reference ......................662

5.250 MHAParser::int_t Class Reference ..........................664

5.251 MHAParser::keyword_list_t Class Reference ....................666

5.252 MHAParser::kw_t Class Reference .........................669

5.253 MHAParser::mcomplex_mon_t Class Reference ..................672

5.254 MHAParser::mcomplex_t Class Reference .....................674

5.255 MHAParser::mﬂoat_mon_t Class Reference ....................676

5.256 MHAParser::mﬂoat_t Class Reference .......................677

5.257 MHAParser::mhaconﬁg_mon_t Class Reference ..................680

5.258 MHAParser::mhapluginloader_t Class Reference .................682

5.259 MHAParser::mint_mon_t Class Reference .....................685

5.260 MHAParser::monitor_t Class Reference .......................686

5.261 MHAParser::parser_t Class Reference .......................688

5.262 MHAParser::range_var_t Class Reference .....................693

5.263 MHAParser::string_mon_t Class Reference ....................696

5.264 MHAParser::string_t Class Reference ........................697

5.265 MHAParser::variable_t Class Reference ......................700

5.266 MHAParser::vcomplex_mon_t Class Reference ..................702

5.267 MHAParser::vcomplex_t Class Reference .....................704

5.268 MHAParser::vﬂoat_mon_t Class Reference .....................707

5.269 MHAParser::vﬂoat_t Class Reference ........................709

5.270 MHAParser::vint_mon_t Class Reference ......................711

5.271 MHAParser::vint_t Class Reference .........................713

5.272 MHAParser::vstring_mon_t Class Reference ....................715

5.273 MHAParser::vstring_t Class Reference .......................717

5.274 MHAParser::window_t Class Reference .......................719

5.275 mhaplug_cfg_t Class Reference ...........................722

5.276 MHAPlugin::cfg_chain_t<runtime_cfg_t >Class Template Reference . . . . . 722

5.277 MHAPlugin::conﬁg_t<runtime_cfg_t >Class Template Reference ........723

5.278 MHAPlugin::plugin_t<runtime_cfg_t >Class Template Reference ........728

5.279 MHAPlugin_Resampling::resampling_if_t Class Reference ............733

5.280 MHAPlugin_Resampling::resampling_t Class Reference .............735

CONTENTS xi

5.281 MHAPlugin_Split::domain_handler_t Class Reference ...............736

5.282 MHAPlugin_Split::dummy_threads_t Class Reference ...............741

5.283 MHAPlugin_Split::posix_threads_t Class Reference ................743

5.284 MHAPlugin_Split::split_t Class Reference ......................747

5.285 MHAPlugin_Split::splitted_part_t Class Reference .................751

5.286 MHAPlugin_Split::thread_platform_t Class Reference ...............756

5.287 MHAPlugin_Split::uni_processor_t Class Reference ................759

5.288 mhaserver_t Class Reference ............................760

5.289 MHASignal::async_rmslevel_t Class Reference ..................763

5.290 MHASignal::delay_spec_t Class Reference .....................765

5.291 MHASignal::delay_t Class Reference ........................766

5.292 MHASignal::delay_wave_t Class Reference ....................768

5.293 MHASignal::doublebuffer_t Class Reference ....................769

5.294 MHASignal::fft_t Class Reference ..........................772

5.295 MHASignal::hilbert_fftw_t Class Reference .....................775

5.296 MHASignal::hilbert_t Class Reference .......................776

5.297 MHASignal::loop_wavefragment_t Class Reference ................778

5.298 MHASignal::matrix_t Class Reference ........................782

5.299 MHASignal::minphase_t Class Reference .....................790

5.300 MHASignal::quantizer_t Class Reference ......................792

5.301 MHASignal::ringbuffer_t Class Reference ......................793

5.302 MHASignal::schroeder_t Class Reference .....................797

5.303 MHASignal::spectrum_t Class Reference ......................800

5.304 MHASignal::stat_t Class Reference .........................805

5.305 MHASignal::subsample_delay_t Class Reference .................806

5.306 MHASignal::uint_vector_t Class Reference .....................809

5.307 MHASignal::waveform_t Class Reference ......................812

5.308 MHASndFile::sf_t Class Reference .........................822

5.309 MHASndFile::sf_wave_t Class Reference ......................823

5.310 MHATableLookup::linear_table_t Class Reference .................824

5.311 MHATableLookup::table_t Class Reference .....................828

5.312 MHATableLookup::xy_table_t Class Reference ...................830

5.313 MHAWindow::bartlett_t Class Reference ......................834

5.314 MHAWindow::base_t Class Reference .......................835

5.315 MHAWindow::blackman_t Class Reference .....................837

5.316 MHAWindow::fun_t Class Reference ........................838

5.317 MHAWindow::hamming_t Class Reference .....................839

5.318 MHAWindow::hanning_t Class Reference .....................840

5.319 MHAWindow::rect_t Class Reference ........................841

5.320 MHAWindow::user_t Class Reference ........................842

5.321 mon_t Class Reference ................................844

5.322 multibandcompressor::fftfb_plug_t Class Reference ................845

5.323 multibandcompressor::interface_t Class Reference ................847

5.324 multibandcompressor::plugin_signals_t Class Reference .............849

5.325 nlms_t Class Reference ...............................851

5.326 noise_t Class Reference ...............................853

5.327 noisePowProposedScale::interface_t Class Reference ..............855

5.328 noisePowProposedScale::noisePowProposed Class Reference .........857

5.329 overlapadd::overlapadd_if_t Class Reference ...................859

5.330 overlapadd::overlapadd_t Class Reference .....................862

5.331 parser_int_dyn Class Reference ...........................864

xii CONTENTS

5.332 plingploing::if_t Class Reference ...........................865

5.333 plingploing::plingploing_t Class Reference .....................868

5.334 plug_t Class Reference ................................870

5.335 plugin_interface_t Class Reference .........................872

5.336 pluginbrowser_t Class Reference ..........................873

5.337 plugindescription_t Class Reference .........................875

5.338 PluginLoader::conﬁg_ﬁle_splitter_t Class Reference ................876

5.339 PluginLoader::fourway_processor_t Class Reference ...............877

5.340 PluginLoader::mhapluginloader_t Class Reference ................880

5.341 pluginloader_t Class Reference ...........................885

5.342 prediction_error Class Reference ..........................886

5.343 prediction_error_conﬁg Class Reference ......................889

5.344 rmslevel_if_t Class Reference ............................892

5.345 rmslevel_t Class Reference .............................894

5.346 route::interface_t Class Reference ..........................895

5.347 route::process_t Class Reference ..........................897

5.348 rt_nlms_t Class Reference ..............................898

5.349 save_spec_t Class Reference ............................901

5.350 save_wave_t Class Reference ............................902

5.351 shadowﬁlter_begin::cfg_t Class Reference .....................903

5.352 shadowﬁlter_begin::shadowﬁlter_begin_t Class Reference ............905

5.353 shadowﬁlter_end::cfg_t Class Reference ......................906

5.354 shadowﬁlter_end::shadowﬁlter_end_t Class Reference ..............908

5.355 sine_cfg_t Struct Reference .............................909

5.356 sine_t Class Reference ................................910

5.357 smoothgains_bridge::overlapadd_if_t Class Reference ..............912

5.358 smoothgains_bridge::smoothspec_wrap_t Class Reference ...........914

5.359 softclip_t Class Reference ..............................916

5.360 softclipper_t Class Reference ............................917

5.361 softclipper_variables_t Class Reference ......................919

5.362 spec2wave_if_t Class Reference ..........................920

5.363 spec2wave_t Class Reference ............................922

5.364 spec_fader_t Class Reference ............................923

5.365 speechnoise_t Class Reference ...........................924

5.366 steerbf Class Reference ...............................926

5.367 steerbf_conﬁg Class Reference ...........................928

5.368 testplugin::ac_parser_t Class Reference ......................930

5.369 testplugin::conﬁg_parser_t Class Reference ....................932

5.370 testplugin::if_t Class Reference ...........................934

5.371 testplugin::signal_parser_t Class Reference ....................936

5.372 timo_AC Class Reference ..............................937

5.373 timo_params Class Reference ............................939

5.374 timoConﬁg Class Reference .............................940

5.375 timoSmooth Class Reference ............................944

5.376 us_t Class Reference .................................947

5.377 wave2spec_if_t Class Reference ..........................949

5.378 wave2spec_t Class Reference ............................951

5.379 wavrec_t Class Reference ..............................953

5.380 wavwriter_t Class Reference .............................955

5.381 windowselector_t Class Reference .........................957

6 File Documentation 960

CONTENTS xiii

6.1 ac2lsl.cpp File Reference ...............................960

6.2 ac2osc.cpp File Reference ..............................960

6.3 ac2wave.cpp File Reference .............................960

6.4 ac_monitor_type.cpp File Reference ........................961

6.5 ac_monitor_type.hh File Reference .........................961

6.6 acConcat_wave.cpp File Reference .........................961

6.7 acConcat_wave.h File Reference ..........................961

6.8 acmon.cpp File Reference ..............................961

6.9 acPooling_wave.cpp File Reference .........................962

6.10 acPooling_wave.h File Reference ..........................962

6.11 acsave.cpp File Reference ..............................962

6.12 acSteer.cpp File Reference ..............................963

6.13 acSteer.h File Reference ...............................963

6.14 acTransform_wave.cpp File Reference .......................964

6.15 acTransform_wave.h File Reference .........................964

6.16 addsndﬁle.cpp File Reference ............................964

6.17 adm.cpp File Reference ...............................965

6.18 adm.hh File Reference ................................966

6.19 altplugs.cpp File Reference ..............................967

6.20 analysemhaplugin.cpp File Reference ........................967

6.21 analysispath.cpp File Reference ...........................967

6.22 auditory_proﬁle.cpp File Reference .........................968

6.23 auditory_proﬁle.h File Reference ...........................968

6.24 browsemhaplugins.cpp File Reference .......................968

6.25 coherence.cpp File Reference ............................969

6.26 combinechannels.cpp File Reference ........................969

6.27 compiler_id.cpp File Reference ...........................970

6.28 compiler_id.hh File Reference ............................970

6.29 complex_ﬁlter.cpp File Reference ..........................970

6.30 complex_ﬁlter.h File Reference ...........................970

6.31 cpuload.cpp File Reference .............................971

6.32 db.cpp File Reference .................................971

6.33 dc.cpp File Reference .................................971

6.34 dc_afterburn.cpp File Reference ...........................972

6.35 dc_afterburn.h File Reference ............................972

6.36 dc_simple.cpp File Reference ............................973

6.37 delay.cpp File Reference ...............................973

6.38 delaysum.cpp File Reference ............................974

6.39 doasvm_classiﬁcation.cpp File Reference .....................974

6.40 doasvm_classiﬁcation.h File Reference .......................974

6.41 doasvm_feature_extraction.cpp File Reference ...................974

6.42 doasvm_feature_extraction.h File Reference ....................975

6.43 doc_appendix.h File Reference ...........................975

6.44 doc_examples.h File Reference ...........................975

6.45 doc_frameworks.h File Reference ..........................975

6.46 doc_general.h File Reference ............................975

6.47 doc_kernel.h File Reference .............................975

6.48 doc_matlab.h File Reference .............................975

6.49 doc_mhamain.h File Reference ...........................975

6.50 doc_parser.h File Reference .............................975

6.51 doc_plugif.cpp File Reference ............................975

xiv CONTENTS

6.52 doc_plugins.h File Reference ............................975

6.53 doc_system.h File Reference ............................975

6.54 doc_toolbox.h File Reference ............................975

6.55 downsample.cpp File Reference ...........................975

6.56 droptect.cpp File Reference .............................976

6.57 example1.cpp File Reference ............................976

6.58 example2.cpp File Reference ............................976

6.59 example3.cpp File Reference ............................976

6.60 example4.cpp File Reference ............................976

6.61 example5.cpp File Reference ............................976

6.62 example6.cpp File Reference ............................977

6.63 fader_spec.cpp File Reference ............................977

6.64 fader_wave.cpp File Reference ...........................977

6.65 fftfbpow.cpp File Reference ..............................978

6.66 fftﬁlterbank.cpp File Reference ............................978

6.67 fshift.cpp File Reference ...............................979

6.68 fshift.hh File Reference ................................979

6.69 fshift_hilbert.cpp File Reference ...........................979

6.70 gain.cpp File Reference ................................979

6.71 gaintable.cpp File Reference .............................980

6.72 gaintable.h File Reference ..............................980

6.73 generatemhaplugindoc.cpp File Reference .....................981

6.74 gtfb_analyzer.cpp File Reference ..........................982

6.75 hann.cpp File Reference ...............................984

6.76 hann.h File Reference ................................985

6.77 identity.cpp File Reference ..............................985

6.78 ifftshift.cpp File Reference ..............................985

6.79 ifftshift.h File Reference ................................986

6.80 iirﬁlter.cpp File Reference ...............................986

6.81 lpc.cpp File Reference ................................986

6.82 lpc.h File Reference ..................................987

6.83 lpc_bl_predictor.cpp File Reference .........................987

6.84 lpc_bl_predictor.h File Reference ..........................987

6.85 lpc_burg-lattice.cpp File Reference .........................988

6.86 lpc_burg-lattice.h File Reference ...........................988

6.87 matrixmixer.cpp File Reference ...........................988

6.88 mha.cpp File Reference ...............................989

6.89 mha.hh File Reference ................................989

6.90 mha_algo_comm.cpp File Reference ........................994

6.91 mha_algo_comm.h File Reference .........................995

6.92 mha_algo_comm.hh File Reference .........................996

6.93 mha_defs.h File Reference ..............................997

6.94 mha_errno.c File Reference .............................998

6.95 mha_errno.h File Reference .............................999

6.96 mha_error.cpp File Reference ............................1000

6.97 mha_error.hh File Reference .............................1001

6.98 mha_event_emitter.h File Reference ........................1002

6.99 mha_events.cpp File Reference ...........................1002

6.100 mha_events.h File Reference ............................1002

6.101 mha_fftfb.cpp File Reference .............................1002

6.102 mha_fftfb.hh File Reference .............................1004

CONTENTS xv

6.103 mha_ﬁfo.cpp File Reference .............................1005

6.104 mha_ﬁfo.h File Reference ..............................1005

6.105 mha_ﬁlter.cpp File Reference ............................1005

6.106 mha_ﬁlter.hh File Reference .............................1006

6.107 mha_generic_chain.cpp File Reference .......................1007

6.108 mha_generic_chain.h File Reference ........................1008

6.109 mha_io_ifc.h File Reference .............................1008

6.110 mha_multisrc.cpp File Reference ..........................1009

6.111 mha_multisrc.h File Reference ............................1010

6.112 mha_os.cpp File Reference .............................1010

6.113 mha_os.h File Reference ...............................1012

6.114 mha_parser.cpp File Reference ...........................1015

6.115 mha_parser.hh File Reference ............................1017

6.116 mha_plugin.hh File Reference ............................1022

6.117 mha_proﬁling.c File Reference ............................1024

6.118 mha_proﬁling.h File Reference ............................1024

6.119 mha_ruby.cpp File Reference ............................1025

6.120 mha_signal.cpp File Reference ...........................1026

6.121 mha_signal.hh File Reference ............................1029

6.122 mha_signal_fft.h File Reference ...........................1038

6.123 mha_tablelookup.cpp File Reference ........................1038

6.124 mha_tablelookup.hh File Reference .........................1038

6.125 mha_tcp.cpp File Reference .............................1039

6.126 mha_tcp.hh File Reference ..............................1040

6.127 mha_toolbox.h File Reference ............................1042

6.128 mha_windowparser.cpp File Reference .......................1042

6.129 mha_windowparser.h File Reference ........................1042

6.130 mhachain.cpp File Reference ............................1043

6.131 mhafw_lib.cpp File Reference ............................1043

6.132 mhafw_lib.h File Reference ..............................1043

6.133 MHAIOalsa.cpp File Reference ...........................1044

6.134 MHAIOFile.cpp File Reference ............................1046

6.135 MHAIOJack.cpp File Reference ...........................1049

6.136 MHAIOJackdb.cpp File Reference ..........................1052

6.137 MHAIOParser.cpp File Reference ..........................1055

6.138 MHAIOPortAudio.cpp File Reference ........................1057

6.139 MHAIOTCP.cpp File Reference ...........................1060

6.140 mhajack.cpp File Reference .............................1064

6.141 mhajack.h File Reference ...............................1064

6.142 mhamain.cpp File Reference .............................1066

6.143 mhapluginloader.cpp File Reference .........................1067

6.144 mhapluginloader.h File Reference ..........................1067

6.145 mhasndﬁle.cpp File Reference ............................1068

6.146 mhasndﬁle.h File Reference .............................1068

6.147 multibandcompressor.cpp File Reference ......................1069

6.148 nlms_wave.cpp File Reference ............................1069

6.149 noise.cpp File Reference ...............................1070

6.150 noisePowProposedScale.cpp File Reference ....................1070

6.151 overlapadd.cpp File Reference ............................1071

6.152 plingploing.cpp File Reference ............................1071

6.153 pluginbrowser.cpp File Reference ..........................1072

xvi CONTENTS

6.154 pluginbrowser.h File Reference ...........................1072

6.155 prediction_error.cpp File Reference .........................1072

6.156 prediction_error.h File Reference ..........................1072

6.157 resampling.cpp File Reference ............................1073

6.158 rmslevel.cpp File Reference .............................1073

6.159 route.cpp File Reference ...............................1073

6.160 save_spec.cpp File Reference ............................1073

6.161 save_wave.cpp File Reference ............................1073

6.162 shadowﬁlter_begin.cpp File Reference .......................1074

6.163 shadowﬁlter_end.cpp File Reference ........................1074

6.164 sine.cpp File Reference ................................1074

6.165 smoothgains_bridge.cpp File Reference ......................1074

6.166 softclip.cpp File Reference ..............................1075

6.167 spec2wave.cpp File Reference ............................1075

6.168 speechnoise.cpp File Reference ...........................1075

6.169 speechnoise.h File Reference ............................1078

6.170 split.cpp File Reference ................................1078

6.171 steerbf.cpp File Reference ..............................1079

6.172 steerbf.h File Reference ...............................1080

6.173 testalsadevice.c File Reference ...........................1080

6.174 testplugin.cpp File Reference ............................1080

6.175 timoconﬁg.cpp File Reference ............................1081

6.176 timoconﬁg.h File Reference .............................1081

6.177 timoSmooth.cpp File Reference ...........................1081

6.178 timosmooth.h File Reference .............................1082

6.179 transducers.cpp File Reference ...........................1082

6.180 upsample.cpp File Reference ............................1083

6.181 wave2spec.cpp File Reference ............................1083

6.182 wavrec.cpp File Reference ..............................1083

6.183 windowselector.cpp File Reference .........................1084

6.184 windowselector.h File Reference ...........................1084

Index 1085

1 Overview 1

1 Overview

The HörTech Open Master Hearing Aid (openMHA), is a development and evaluation software

platform that is able to execute hearing aid signal processing in real-time on standard computing

hardware with a low delay between sound input and output.

1.1 Structure

The openMHA can be split into four major components :

•The openMHA command line application (MHA) (p. 34)

•Signal processing plugins (p. 7)

• Audio input-output (IO) plugins (see io_ﬁle_t (p. 365), MHAIOJack (p. 98), io_parser_t

(p. 372), io_tcp_parser_t (p. 378))

•The openMHA toolbox library (p. 35)

openMHA

plugins IO

audio backend

(Jack, File, TCP)

MHAlibopenmha

control applications

(e.g., Octave)

Figure 1 openMHA structure

The openMHA command line application (MHA) (p. 34) acts as a plugin host. It can load

signal processing plugins as well as audio input-output (IO) plugins. Additionally, it provides

the command line conﬁguration interface and a TCP/IP based conﬁguration interface. Several

IO plugins exist: For real-time signal processing, commonly the openMHA MHAIOJack (p. 98)

plugin (see plugins' manual) is used, which provides an interface to the Jack Audio Connection

Kit (JACK). Other IO plugins provide audio ﬁle access or TCP/IP-based processing.

openMHA plugins (p. 7) provide the audio signal processing capabilities and audio signal han-

dling. Typically, one openMHA plugin implements one speciﬁc algorithm. The complete virtual

hearing aid signal processing can be achieved by a combination of several openMHA plugins.

2 CONTENTS

1.2 Platform Services and Conventions

The openMHA platform offers some services and conventions to algorithms implemented in

plugins, that make it especially well suited to develop hearing aid algorithms, while still support-

ing general-purpose signal processing.

1.2.1 Audio Signal Domains

As in most other plugin hosts, the audio signal in the openMHA is processed in audio chunks.

However, plugins are not restricted to propagate audio signal as blocks of audio samples in the

time domain another option is to propagate the audio signal in the short time Fourier transform

(STFT) domain, i.e. as spectra of blocks of audio signal, so that not every plugin has to perform

its own STFT analysis and synthesis. Since STFT analysis and re-synthesis of acceptable

audio quality always introduces an algorithmic delay, sharing STFT data is a necessity for a

hearing aid signal processing platform, because the overall delay of the complete processing

has to be as short as possible.

Similar to some other platforms, the openMHA allows also arbitrary data to be exchanged

between plugins through a mechanism called algorithm communication variables (p. 28) or

short "AC vars". This mechanism is commonly used to share data such as ﬁlter coefﬁcients or

ﬁlter states.

1.2.2 Real-Time Safe Complex Conﬁguration Changes

Hearing aid algorithms in the openMHA can export conﬁguration settings that may be changed

by the user at run time.

To ensure real-time safe signal processing, the audio processing will normally be done in a

signal processing thread with real-time priority, while user interaction with conﬁguration pa-

rameters would be performed in a conﬁguration thread with normal priority, so that the audio

processing does not get interrupted by conﬁguration tasks. Two types of problems may occur

when the user is changing parameters in such a setup:

• The change of a simple parameter exposed to the user may cause an involved recalcu-

lation of internal runtime parameters that the algorithm actually uses in processing. The

duration required to perform this recalculation may be a signiﬁcant portion of (or take even

longer than) the time available to process one block of audio signal. In hearing aid usage,

it is not acceptable to halt audio processing for the duration that the recalculation may

require.

• If the user needs to change multiple parameters to reach a desired conﬁguration state

of an algorithm from the original conﬁguration state, then it may not be acceptable that

processing is performed while some of the parameters have already been changed while

others still retain their original values. It is also not acceptable to interrupt signal process-

ing until all pending conﬁguration changes have been performed.

1.2 Platform Services and Conventions 3

The openMHA provides a mechanism in its toolbox library to enable real-time safe conﬁguration

changes in openMHA plugins:

Basically, existing runtime conﬁgurations are used in the processing thread until the work of

creating an updated runtime conﬁguration has been completed in the conﬁguration thread.

In hearing aids, it is more acceptable to continue to use an outdated conﬁguration for a few

more milliseconds than blocking all processing.

The openMHA toolbox library provides an easy-to-use mechanism to integrate real-time safe

runtime conﬁguration updates into every plugin.

1.2.3 Plugins can Themselves Host Other Plugins

An openMHA plugin can itself act as a plugin host. This allows to combine analysis and re-

synthesis methods in a single plugin. We call plugins that can themselves load other plugins

“bridge plugins'' in the openMHA.

When such a bridge plugin is then called by the openMHA to process one block of signal, it

will ﬁrst perform its analysis, then invoke (as a function call) the signal processing in the loaded

plugin to process the block of signal in the analysis domain, wait to receive a processed block of

signal in the analysis domain back from the loaded plugin when the signal processing function

call to that plugin returns, then perform the re-synthesis transform, and ﬁnally return the block

of processed signal in the original domain back to the caller of the bridge plugin.

1.2.4 Central Calibration

The purpose of hearing aid signal processing is to enhance the sound for hearing impaired

listeners. Hearing impairment generally means that people suffering from it have increased

hearing thresholds, i.e. soft sounds that are audible for normal hearing listeners may be im-

perceptible for hearing impaired listeners. To provide accurate signal enhancement for hearing

impaired people, hearing aid signal processing algorithms have to be able to determine the

absolute physical sound pressure level corresponding to a digital signal given to any openM←-

HA plugin for processing. Inside the openMHA, we achieve this with the following convention:

The single-precision ﬂoating point time-domain sound signal samples, that are processed in-

side the openMHA plugins in blocks of short durations, have the physical pressure unit Pascal

(1Pa = 1N/m2). With this convention in place, all plugins can determine the absolute phys-

ical sound pressure level from the sound samples that they process. A derived convention is

employed in the spectral domain for STFT signals. Due to the dependency of the calibration

on the hardware used, it is the responsibility of the user of the openMHA to perform calibration

measurements and adapt the openMHA settings to make sure that this calibration convention is

met. We provide the plugin transducers which can be conﬁgured to perform the necessary

signal adjustments.

4 CONTENTS

2 Todo List

Class AuditoryProﬁle::proﬁle_t (p. 226)

Give more documentation; implement all parts of the auditory proﬁle.

Class mha_stash_environment_variable_t (p. 466)

Move to collection of unit-test support classes when we have one.

Class mhaconﬁg_t (p. 504)

Add information on number of bands and on center frequencies, or replace by mha_audio←-

_descriptor_t (p. 427).

Class MHAFilter::ﬁlter_t (p. 531)

Implement a more robust ﬁlter form.

Member MHAFilter::polyphase_resampling_t::now_index (p. 563)

Index into what? What is the meaning of now?

Class MHAPlugin::plugin_t<runtime_cfg_t >(p. 728)

Describe all services provided by this class, so that the reason why it is recommended that

all plugins use this class as their base is evident. Document all relevant methods and ﬁelds.

3 Module Documentation

3.1 Concept of Variables and Data Exchange in the openMHA

Accessibility of conﬁguration variables and data exchange between plugins (processing blocks)

are an important issue in the openMHA.

Accessibility of conﬁguration variables and data exchange between plugins (processing blocks)

are an important issue in the openMHA.

In general, variable types in the openMHA are distinguished by their different access methods.

The variable types in the openMHA are:

•Conﬁguration variables : Read and write accesses are possible through the openM←-

HA conﬁguration language interface. Conﬁguration variables are implemented as C++

classes with a public data member of the underlying C type. Conﬁguration variables

can be read and modiﬁed from “outside'' using the conﬁguration language. The plugin

which provides the conﬁguration variable can use the exposed data member directly. All

accesses through the openMHA conﬁguration language are checked for data type, valid

range, and access restrictions.

•Monitor variables : Read access is possible through the openMHA conﬁguration lan-

guage. Write access is only possible from the C++ code. Internally, monitor variables

have a similar C++ class interface as conﬁguration variables.

3.1 Concept of Variables and Data Exchange in the openMHA 5

•AC variables (algorithm communication variables (p. 28)): Any C or C++ data struc-

ture can be shared within an openMHA chain. Access management and name space

is realised in openMHA chain plugin ('mhachain'). AC variables are not available to the

openMHA conﬁguration language interface, although a read-only converter plugin acmon

is available.

•Runtime conﬁguration : Algorithms usually derive more parameters (runtime conﬁgura-

tion) from the openMHA conﬁguration language variables. When a conﬁguration variable

changes through conﬁguration language write access, then the runtime conﬁguration has

to be recomputed. Plugin developers are encouraged to encapsulate the runtime con-

ﬁguration in a C++ class, which recomputes the runtime conﬁguration from conﬁguration

variables in the constructor. The openMHA supports lock-free and thread-safe replace-

ment of the runtime conﬁguration instance (see example5.cpp (p. 21) and references

therein).

Plugin specific Chain specific

variables

monitors

runtime configuation

(e.g. C++ class)

algorithm communication

(AC variables)

Configuration language

Figure 2 Variable types in the openMHA

The C++ data types are shown in the ﬁgure below. These variables can be accessed via the

openMHA host application using the openMHA conﬁguration language. For more details see

'Application engineers' manual'.

6 CONTENTS

MHAParser elements

Monitors Variables parser_t

kw_t

string_t

vstring_t

bool_t

int_t

float_t

vfloat_t

mfloat_t

Numeric variables

string_mon_t

vstring_mon_t

int_mon_t

float_mon_t

vfloat_mon_t

vint_tvint_mon_t

Numeric monitors

complex_t

vcompelx_t

complex_mon_t

vcomplex_mon_t

mfloat_mon_t

Figure 3 MHAParser elements

3.2 The openMHA Plugins (programming interface) 7

3.2 The openMHA Plugins (programming interface)

An openMHA plugin is the signal processing unit, usually an algorithm.

Classes

• class MHAPlugin::plugin_t<runtime_cfg_t >

The template class for C++ openMHA plugins.

Macros

• #deﬁne MHAPLUGIN_CALLBACKS_PREFIX(preﬁx, classname, indom, outdom)

C++ wrapper macro for the plugin interface.

• #deﬁne MHAPLUGIN_CALLBACKS(plugname, classname, indom, outdom) MHAPLU←-

GIN_CALLBACKS_PREFIX(MHA_STATIC_ ## plugname ## _,classname,indom,outdom)

C++ wrapper macro for the plugin interface.

• #deﬁne MHAPLUGIN_DOCUMENTATION(plugname, cat, doc) MHAPLUGIN_DOCU←-

MENTATION_PREFIX(MHA_STATIC_ ## plugname ## _,cat,doc)

Wrapper macro for the plugin documentation interface.

3.2.1 Detailed Description

An openMHA plugin is the signal processing unit, usually an algorithm.

openMHA plugins can be combined into processing chains. One of the conﬁgured chains can

be selected for output which allows direct comparison of single algorithms or complex signal

processing conﬁgurations. Algorithms within one chain can communicate with each other by

sharing some of their variables, see section Communication between algorithms (p. 28).

The openMHA plugins can use the openMHA conﬁguration language for their conﬁguration.

If they do so, the conﬁguration can be changed through the framework even at run time. A

description of this language can be found in section The openMHA conﬁguration language

(p. 34). If the algorithms should make use of the openMHA conﬁguration language, they need

to be written in C++ rather than pure C.

In the openMHA package a set of example plugins is included. These examples are the base

of a step by step tutorial on how to write an openMHA plugin. See section Writing openMHA

Plugins. A step-by-step tutorial (p. 11) for details.

openMHA plugins communicate with the openMHA using a simple C language interface. This

way it is easy to mix plugins compiled with different C++ compilers. For convenience, we

provide C++ classes which can be connected to the C++ interface. We strongly recommend

the usage of these C++ wrappers. They include out-of-the box support exporting variables to

the conﬁguration interface and for thread safe conﬁguration update.

The openMHA C++ plugin interface consists of a few number of method prototypes:

The output domain (spectrum or waveform) of an openMHA plugin will typically be the same as

the input domain:

8 CONTENTS

•mha_wave_t (p. 496)∗process(mha_wave_t (p. 496)∗): pure waveform processing

•mha_spec_t (p. 465)∗process(mha_spec_t (p. 465)∗): pure spectral processing

But it is also possible to implement domain transformations (from the time domain into spectrum

or vice versa). The corresponding method signatures are:

•mha_spec_t (p. 465)∗process(mha_wave_t (p. 496)∗): Domain transformation from

waveform to spectrum

•mha_wave_t (p. 496)∗process(mha_spec_t (p. 465)∗): Domain transformation from

spectrum to waveform

For preparation and release of a plugin, the methods

• void prepare(mhaconﬁg_t (p. 504) &) and

• void release(void)

have to be implemented. The openMHA will call the process() method only ater the pre-

pare method has returned and before release() is invoked. It is guarantteed by the open←-

MHA framework that signal processing is performed only between calls of prepare() and

release(). Each call of prepare() is followed by a call of release() (after some optional

signal processing).

For conﬁguration purposes, the plugin class has to export a method called parse() which

implements the openMHA conﬁguration language. We strongly recommend that you do not

implement this method yourself, but by inheriting from the class MHAParser::parser_t (p. 688)

from the openMHA toolbox, directly or indirectly (inheriting from a class that itself inherits from

MHAParser::parser_t (p. 688)).

3.2.2 Connecting the C++ class with the C Interface

A C++ class which provides the appropriate methods can be used as an openMHA Plugin by

connecting it to the C interface using the MHAPLUGIN_CALLBACKS (p. 10) macro.

The openMHA Toolbox library provides a base class MHAPlugin::plugin_t (p. 728)<T>(a

template class) which can be used as the base class for a plugin class. This base class im-

plements some necessary features for openMHA plugin developers like integration into the

openMHA conﬁguration language environment (it inherits from MHAParser::parser_t (p. 688))

and thread-safe runtime conﬁguration update.

3.2 The openMHA Plugins (programming interface) 9

3.2.3 Error reporting

When your plugin detects a situation that it cannot handle, like input signal of the wrong signal

domain at preparation time, unsupported number of input channels at preparation time, unsup-

ported combinations of values in the plugin's variables during conﬁguration, it should throw a

C++ exception. The exception should be of type MHAError. Exceptions of this type are caught

by the MHAPLUGIN_CALLBACKS (p. 10) macro for further error Reporting.

Throwing exceptions in response to unsupported conﬁguration changes does not stop the sig-

nal processing. The openMHA conﬁguration language parser will restore the previous value

of that variable and report an error to the conﬁgurator, while the signal processing continues.

Throwing exceptions from the signal processing thread will terminate the signal processing.

Therefore, you should generally avoid throwing exceptions from the process method. Only do

this if you detected a defect in your plugin, and then you should include enough information in

the error message to be able to ﬁx the defect.

3.2.4 Contents of the openMHA Plugin programming interface

3.2.5 Macro Deﬁnition Documentation

3.2.5.1 #deﬁne MHAPLUGIN_CALLBACKS_PREFIX(

preﬁx,

classname,

indom,

outdom )

C++ wrapper macro for the plugin interface.

Parameters

classname The name of the plugin class

indom Input domain (wave or spec)

outdom Output domain (wave or spec)

This macro deﬁnes all required openMHA Plugin interface functions and passes calls of these

functions to the corresponding member functions of the class ‘classname'. The parameters

‘indom' and ‘outdom' specify the input and output domain of the processing method. The

MHAInit() and MHADestroy() functions will create or destroy an instance of the class. The

approriate member functions have to be deﬁned in the class. It is suggested to make usage of

the MHAPlugin::plugin_t (p. 728) template class. Exceptions of type MHA_Error (p. 445) are

caught and transformed into apropriate error codes with their corresponding error messages.

10 CONTENTS

3.2.5.2 #deﬁne MHAPLUGIN_CALLBACKS(

plugname,

classname,

indom,

outdom )MHAPLUGIN_CALLBACKS_PREFIX(MHA_STATIC_ ## plugname

## _,classname,indom,outdom)

C++ wrapper macro for the plugin interface.

Parameters

plugname The ﬁle name of the plugin without the .so or .dll extension

classname The name of the plugin class

indom Input domain (wave or spec)

outdom Output domain (wave or spec)

This macro deﬁnes all required openMHA Plugin interface functions and passes calls of these

functions to the corresponding member functions of the class ‘classname'. The parameters

‘indom' and ‘outdom' specify the input and output domain of the processing method. The MHA←-

Init() and MHADestroy() functions will create or destroy an instance of the class. The approriate

member functions have to be deﬁned in the class. It is suggested to make usage of the MH←-

APlugin::plugin_t (p. 728) template class. Exceptions of type MHA_Error (p. 445) are caught

and transformed into apropriate error codes with their corresponding error messages.

3.2.5.3 #deﬁne MHAPLUGIN_DOCUMENTATION(

plugname,

cat,

doc )MHAPLUGIN_DOCUMENTATION_PREFIX(MHA_STATIC_ ##

plugname ## _,cat,doc)

Wrapper macro for the plugin documentation interface.

Parameters

plugin The ﬁle name of the plugin without the .so or .dll extension

cat Space separated list of categories to which belong the plugin (as const char∗)

doc Documentation of the plugin (as const char∗)

This macro deﬁnes the openMHA Plugin interface function for the documentation. The cate-

gories can be any space seperated list of category names. An empty string will categorize the

plugin in the category 'other'.

The documentation should contain a description of the plugin including a description of the

underlying models, and a paragraph containing hints for usage. The text should be LaTeX

compatible (e.g., avoid or quote underscores in the text part); equations should be formatted

as LaTeX.

3.3 Writing openMHA Plugins. A step-by-step tutorial 11

3.3 Writing openMHA Plugins. A step-by-step tutorial

A step-by-step tutorial on writing openMHA plugins.

openMHA contains a small number of example plugins as C++ source code. They are meant to

help developers in understanding the concepts of openMHA plugin programming starting from

the simplest example and increasing in complexity. This tutorial explains the basic parts of the

example ﬁles.

3.3.1 example1.cpp

The example plugin ﬁle example1.cpp (p. 976)demonstrates the easiest way to implement

an openMHA Plugin. It attenuates the sound signal in the ﬁrst channel by multiplying the

sound samples with a factor. The plugin class MHAPlugin::plugin_t (p. 728) exports several

methods, but only two of them need a non-empty implementation: prepare() method is a

pure virtual function and process() is called when signal processing starts.

#include "mha_plugin.hh"

class example1_t : public MHAPlugin::plugin_t<int> {

public:

example1_t(algo_comm_t & ac,

const std::string & chain_name,

const std::string & algo_name)

: MHAPlugin::plugin_t<int>("",ac)

{/*Do nothing in constructor */}

void release(void)

{/*Do nothing in release */}

Every plugin implementation should include the 'mha_plugin.hh (p. 1022)' header ﬁle. C++

helper classes for plugin development are declared in this header ﬁle, and most header ﬁles

needed for plugin development are included by mha_plugin.hh (p. 1022).

The class plugin1_t inherits from the class MHAPlugin::plugin_t (p. 728), which then inher-

its from MHAParser::parser_t (p. 688) – the conﬁguration language interface in the method

"parse". Our plugin class therefore exports the working "parse" method inherited from MHA←-

Parser::parser_t (p. 688), and the plugin is visible in the openMHA conﬁguration tree.

The constructor has to accept 3 parameters of correct types. In this simple example, we do not

make use of them.

The release() method is used to free resources after signal processing. In this simple ex-

ample, we do not allocate resources, so there is no need to free them.

3.3.1.1 The prepare method

void prepare(mhaconfig_t & signal_info)

{

if (signal_info.domain != MHA_WAVEFORM)

throw MHA_Error(__FILE__, __LINE__,

"This plugin can only process waveform signals.");

if (signal_info.channels < 1)

throw MHA_Error(__FILE__,__LINE__,

"This plugin requires at least one input channel.");

}

12 CONTENTS

Parameters

signal_info Contains information about the input signal's parameters, see mhaconﬁg_t

(p. 504).

The prepare() method of the plugin is called before the signal processing starts, when the

input signal parameters like domain, number of channels, frames per block, and sampling rate

are known. The prepare() method can check these values and raise an exception if the

plugin cannot cope with them, as is done here. The plugin can also change these values if the

signal processing performed in the plugin results in an output signal with different parameters.

This plugin does not change the signal's parameters, therefore they are not modiﬁed here.

3.3.1.2 The signal processing method

mha_wave_t *process(mha_wave_t *signal)

{

unsigned int channel = 0; // channels and frames counting starts with 0

float factor = 0.1f;

unsigned int frame;

// Scale channel number "channel" by "factor":

for(frame = 0; frame < signal->num_frames; frame++) {

// Waveform channels are stored interleaved.

signal->buf[signal->num_channels *frame + channel] *= factor;

}

// Algorithms may process data in-place and return the input signal

// structure as their output signal:

return signal;

}

};

Parameters

signal Pointer to the input signal structure mha_wave_t (p. 496).

Returns

Pointer to the output signal structure. The input signal structure may be reused if the

signal has the same domain and dimensions.

The plugin works with time domain input signal (indicated by the data type mha_wave_t (p. 496)

of the process method's parameter). It scales the ﬁrst channel by a factor of 0.1. The output

signal reuses the structure that previously contained the input signal (in-place processing).

3.3.1.3 Connecting the C++ class with the C plugin interface

Plugins have to export C functions as their interface (to avoid C++ name-mangling issues and

other incompatibilities when mixing plugins compiled with different C++ compilers).

MHAPLUGIN_CALLBACKS(example1,example1_t,wave,wave)

3.3 Writing openMHA Plugins. A step-by-step tutorial 13

This macro takes care of accessing the C++ class from the C functions required as the plugin's

interface. It implements the C funtions and calls the corresponding C++ instance methods.

Plugin classes should be derived from the template class MHAPlugin::plugin_t (p. 728) to be

compatible with the C interface wrapper.

This macro also catches C++ exceptions of type MHA_Error (p. 445), when raised in the meth-

ods of the plugin class, and reports the error using an error ﬂag as the return value of the

underlying C function. It is therefore important to note that only C++ exceptions of type MH←-

A_Error (p. 445) may be raised by your plugin. If your code uses different Exception classes,

you will have to catch them yourself before control leaves your plugin class, and maybe report

the error by throwing an instance of MHA_Error (p. 445). This is important, because: (1) C++

exceptions cannot cross the plugin interface, which is in C, and (2) there is no error handling

code for your exception classes in the openMHA framework anyways.

3.3.2 example2.cpp

This is another simple example of openMHA plugin written in C++. This plugin also scales one

channel of the input signal, working in the time domain. The scale factor and which channel to

scale (index number) are made accessible to the conﬁguration language.

The algorithm is again implemented as a C++ class.

class example2_t : public MHAPlugin::plugin_t<int> {

MHAParser::int_t scale_ch;

MHAParser::float_t factor;

public:

example2_t(algo_comm_t & ac,

const std::string & chain_name,

const std::string & algo_name);

void prepare(mhaconfig_t & signal_info);

void release(void);

mha_wave_t *process(mha_wave_t *signal);

};

Parameters

scale_ch – the channel number to be scaled

factor – the scale factor of the scaling.

This class again inherits from the template class MHAPlugin::plugin_t (p. 728) for intergration

with the openMHA conﬁguration language. The two data members serve as externally visible

conﬁguration variables. All methods of this class have a non-empty implementation.

3.3.2.1 Constructor

example2_t::example2_t(algo_comm_t & ac,

const std::string & chain_name,

14 CONTENTS

const std::string & algo_name)

: MHAPlugin::plugin_t<int>("This plugin multiplies the sound signal"

" in one audio channel by a factor",ac),

scale_ch("Index of audio channel to scale. Indices start from 0.",

"0",

"[0,["),

factor("The scaling factor that is applied to the selected channel.",

"0.1",

"[0,[")

{

insert_item("channel", &scale_ch);

insert_item("factor", &factor);

}

The constructor invokes the superclass constructor with a string parameter. This string pa-

rameter serves as the help text that describes the functionality of the plugin. The constructor

registers conﬁguration variables with the openMHA conﬁguration tree and sets their default val-

ues and permitted ranges. The minimum permitted value for both variables is zero, and there is

no maximum limit (apart from the limitations of the underlying C data type). The conﬁguration

variables have to be registered with the parser node instance using the MHAParser::parser←-

_t::insert_item (p. 690) method.

3.3.2.2 The prepare method

void example2_t::prepare(mhaconfig_t & signal_info)

{

if (signal_info.domain != MHA_WAVEFORM)

throw MHA_Error(__FILE__, __LINE__,

"This plugin can only process waveform signals.");

// The user may have configured scale_ch before prepare is called.

// Check that the configured channel is present in the input signal.

if (signal_info.channels <= unsigned(scale_ch.data))

throw MHA_Error(__FILE__,__LINE__,

"This plugin requires at least %d input channels.",

scale_ch.data + 1);

// Adjust the range of the channel configuration variable so that it

// cannot be set to an out-of-range value during processing.

using MHAParser::StrCnv::val2str;

scale_ch.set_range("[0," + val2str(int(signal_info.channels)) + "[");

}

Parameters

signal_info – contains information about the input signal's parameters, see mhaconﬁg_t

(p. 504).

The user may have changed the conﬁguration variables before preparing the openMHA plugin.

A consequence of this is that it is not sufﬁcient any more to check if the input signal has at least

1 audio channel.

Instead, this prepare method checks that the input signal has enough channels so that the cur-

rent value of scale_ch.data is a valid channel index, i.e. 0 ≤scale_ch.data <signal←-

_info.channels. The prepare method does not have to check that 0 ≤scale_ch.data,

since this is guaranteed by the valid range setting of the conﬁguration variable.

3.3 Writing openMHA Plugins. A step-by-step tutorial 15

The prepare method then modiﬁes the valid range of the scale_ch variable, it modiﬁes the up-

per bound so that the user cannot set the variable to a channel index higher than the available

channels. Setting the range is done using a string parameter. The prepare method contate-

nates a string of the form "[0,n[". n is the number of channels in the input signal, and is used

here as an exclusive upper boundary. To convert the number of channels into a string, a helper

function for string conversion from the openMHA Toolbox is used. This function is overloaded

and works for several data types.

It is safe to assume that the value of conﬁguration variables does not change while the pre-

pare method executes, since openMHA preparation is triggered from a conﬁguration language

command, and the openMHA conﬁguration language parser is busy and cannot accept other

commands until all openMHA plugins are prepared (or one of them stops the process by raising

an exception). As we will see later in this tutorial, the same assumption cannot be made for the

process method.

3.3.2.3 The release method

void example2_t::release(void)

{

scale_ch.set_range("[0,[");

}

The release method should undo the state changes that were performed by the prepare

method. In this example, the prepare method has reduced the valid range of the scale_←-

ch, so that only valid channels could be selected during signal processing.

The release method reverts this change by setting the valid range back to its original value,

"[0,[".

3.3.2.4 The signal processing method

mha_wave_t *example2_t::process(mha_wave_t *signal)

{

unsigned int frame;

for(frame = 0; frame < signal->num_frames; frame++)

value(signal,frame,scale_ch.data) *= factor.data;

return signal;

}

The processing function uses the current values of the conﬁguration variables to scale every

frame in the selected audio channel.

Note that the value of each conﬁguration variable can change while the processing method

executes, since the process method usually executes in a different thread than the conﬁguration

interface.

For this simple plugin, this is not a problem, but for more advanced plugins, it has to be taken

into consideration. The next section takes a closer look at the problem.

16 CONTENTS

Consistency

Assume that one thread reads the value stored in a variable while another thread writes a new

value to that variable concurrently. In this case, you may have a consistency problem. You

would perhaps expect that the value retrieved from the variable either (a) the old value, or (b)

the new value, but not (c) something else. Yet generally case (c) is a possibility.

Fortunately, for some data types on PC systems, case (c) cannot happen. These are 32bit

wide data types with a 4-byte alignment. Therefore, the values in MHAParser::int_t (p. 664)

and MHAParser::ﬂoat_t (p. 659) are always consistent, but this is not the case for vectors,

strings, or complex values. With these, you can get a mixture of the bit patterns of old and new

values, or you can even cause a memory access violation in case a vector or string grows and

has to be reallocated to a different memory address.

There is also a consistency problem if you take the combination of two "safe" datatypes. The

openMHA provides a mechanism that can cope with these types of problems. This thread-safe

runtime conﬁguration update mechanism is introduced in example 5.

3.3.3 example3.cpp

This example introduces the openMHA Event mechanism. Plugins that provide conﬁguration

variable can receive a callback from the parser base class when a conﬁguration variable is

accessed through the conﬁguration language interface.

The third example performes the same processing as before, but now only even channel indices

are permitted when selecting the audio channel to scale. This restriction cannot be ensured by

setting the range of the channel index conﬁguration variable. Instead, the event mechanism of

openMHA conﬁguration variables is used. Conﬁguration variables emit 4 different events, and

your plugin can connect callback methods that are called when the events are triggered. These

events are:

writeaccess

• triggered on write access to a conﬁguration variable.

valuechanged

• triggered when write access to a conﬁguration variable actually changes the value of this

variable.

readaccess

• triggered after the value of the conﬁguration variable has been read.

prereadaccess

• triggered before the value of a conﬁguration variable is read, i.e. the value of the re-

quested variable can be changed by the callback to implement computation on demand.

All of these callbacks are executed in the conﬁguration thread. Therefore, the callback imple-

mentation does not have to be realtime-safe. No other updates of conﬁguration language vari-

ables through the conﬁguration language can happen in parallel, but your processing method

can execute in parallel and may change values.

3.3 Writing openMHA Plugins. A step-by-step tutorial 17

3.3.3.1 Data member declarations

class example3_t : public MHAPlugin::plugin_t<int> {

MHAParser::int_t scale_ch;

MHAParser::float_t factor;

MHAParser::int_mon_t prepared;

MHAEvents::patchbay_t<example3_t> patchbay;

This plugin exposes another conﬁguration variable, "prepared", that keeps track of the pre-

pared state of the plugin. This is a read-only (monitor) integer variable, i.e. its value can only

be changed by your plugin's C++ code. When using the conﬁguration language interface, the

value of this variable can only be read, but not changed.

The patchbay member is an instance of a connector class that connects event sources with

callbacks.

3.3.3.2 Method declarations

/*Callbacks triggered by Events */

void on_scale_ch_writeaccess();

void on_scale_ch_valuechanged();

void on_scale_ch_readaccess();

void on_prereadaccess();

public:

example3_t(algo_comm_t & ac,

const std::string & chain_name,

const std::string & algo_name);

void prepare(mhaconfig_t & signal_info);

void release(void);

mha_wave_t *process(mha_wave_t *signal);

};

This plugin exposes 4 callback methods that are triggered by events. Multiple events (from the

same or different conﬁguration variables) can be connected to the same callback method, if

desired.

This example plugin uses the valuechanged event to check that the scale_ch conﬁguration

variable is only set to valid values.

The other callbacks only cause log messages to stdout, but the comments in the logging call-

backs give a hint when listening on the events would be useful.

18 CONTENTS

3.3.3.3 Example 3 constructor

example3_t::example3_t(algo_comm_t & ac,

const std::string & chain_name,

const std::string & algo_name)

: MHAPlugin::plugin_t<int>("This plugin multiplies the sound signal"

" in one audio channel by a factor",ac),

scale_ch("Index of audio channel to scale. Indices start from 0."

" Only channels with even indices may be scaled.",

"0",

"[0,["),

factor("The scaling factor that is applied to the selected channel.",

"0.1",

"[0,["),

prepared("State of this plugin: 0 = unprepared, 1 = prepared")

{

insert_item("channel", &scale_ch);

insert_item("factor", &factor);

prepared.data = 0;

insert_item("prepared", &prepared);

patchbay.connect(&scale_ch.writeaccess, this,

&example3_t::on_scale_ch_writeaccess);

patchbay.connect(&scale_ch.valuechanged, this,

&example3_t::on_scale_ch_valuechanged);

patchbay.connect(&scale_ch.readaccess, this,

&example3_t::on_scale_ch_readaccess);

patchbay.connect(&scale_ch.prereadaccess, this,

&example3_t::on_prereadaccess);

patchbay.connect(&factor.prereadaccess, this,

&example3_t::on_prereadaccess);

patchbay.connect(&prepared.prereadaccess, this,

&example3_t::on_prereadaccess);

}

The constructor of monitor variables does not take a parameter for setting the initial value.

The single parameter here is the help text describing the contents of the read-only variable.

If the initial value should differ from 0, then the .data member of the conﬁguration variable

has to be set to the initial value in the plugin constructor's body explicitly, as is done here for

demonstration although the initial value of this monitor variable is 0.

Events and callback methods are then connected using the patchbay member variable.

3.3.3.4 The prepare method

void example3_t::prepare(mhaconfig_t & signal_info)

{

if (signal_info.domain != MHA_WAVEFORM)

throw MHA_Error(__FILE__, __LINE__,

"This plugin can only process waveform signals.");

// The user may have configured scale_ch before prepare is called.

// Check that the configured channel is present in the input signal.

if (signal_info.channels <= unsigned(scale_ch.data))

throw MHA_Error(__FILE__,__LINE__,

"This plugin requires at least %d input channels.",

scale_ch.data + 1);

// bookkeeping

prepared.data = 1;

}

The prepare method checks wether the current setting of the scale_ch variable is possible with

the input signal dimension. It does not adjust the range of the variable, since the range alone

is not sufﬁcient to ensure all future settings are also valid: The scale channel index has to be

even.

3.3 Writing openMHA Plugins. A step-by-step tutorial 19

3.3.3.5 The release method

void example3_t::release(void)

{

prepared.data = 0;

}

The release method is needed for tracking the prepared state only in this example.

3.3.3.6 The signal processing method

mha_wave_t *example3_t::process(mha_wave_t *signal)

{

unsigned int frame;

for(frame = 0; frame < signal->num_frames; frame++)

value(signal,frame,scale_ch.data) *= factor.data;

return signal;

}

The signal processing member function is the same as in example 2.

3.3.3.7 The callback methods

void example3_t::on_scale_ch_writeaccess()

{

printf("Write access: Attempt to set scale_ch=%d.\n", scale_ch.data);

// Can be used to track any writeaccess to the configuration, even

// if it does not change the value. E.g. setting the name of the

// sound file in a string configuration variable can cause a sound

// file player plugin to start playing the sound file from the

// beginning.

}

void example3_t::on_scale_ch_valuechanged()

{

if (scale_ch.data & 1)

throw MHA_Error(__FILE__,__LINE__,

"Attempt to set scale_ch to non-even value %d",

scale_ch.data);

// Can be used to recompute a runtime configuration only if some

// configuration variable actually changed.

}

void example3_t::on_scale_ch_readaccess()

{

printf("scale_ch has been read.\n");

// A configuration variable used as an accumulator can be reset

// after it has been read.

}

void example3_t::on_prereadaccess()

{

printf("A configuration language variable is about to be read.\n");

// Can be used to compute the value on demand.

}

MHAPLUGIN_CALLBACKS(example3,example3_t,wave,wave)

When the writeaccess or valuechanged callbacks throw an MHAError exception, then the

change made to the value of the conﬁguration variable is reverted.

If multiple event sources are connected to a single callback method, then it is not possible

to determine which event has caused the callback to execute. Often, this information is not

crucial, i.e. when the answer to a change of any variable in a set of variables is the same, e.g.

the recomputation of a new runtime conﬁguration that takes all variables of this set as input.

20 CONTENTS

3.3.4 example4.cpp

This plugin is the same as example 3 except that it works on the spectral domain (STFT).

3.3.4.1 The Prepare method

void example4_t::prepare(mhaconfig_t & signal_info)

{

if (signal_info.domain != MHA_SPECTRUM)

throw MHA_Error(__FILE__, __LINE__,

"This plugin can only process spectrum signals.");

// The user may have configured scale_ch before prepare is called.

// Check that the configured channel is present in the input signal.

if (signal_info.channels <= unsigned(scale_ch.data))

throw MHA_Error(__FILE__,__LINE__,

"This plugin requires at least %d input channels.",

scale_ch.data + 1);

// bookkeeping

prepared.data = 1;

}

The prepare method now checks that the signal domain is MHA_SPECTRUM.

3.3.4.2 The signal processing method

mha_spec_t *example4_t::process(mha_spec_t *signal)

{

unsigned int bin;

// spectral signal is stored non-interleaved.

mha_complex_t *channeldata =

signal->buf + signal->num_frames *scale_ch.data;

for(bin = 0; bin < signal->num_frames; bin++)

channeldata[bin] *= factor.data;

return signal;

}

The signal processing member function works on the spectral signal instead of the wave signal

as before.

The mha_spec_t (p. 465) instance stores the complex (mha_complex_t (p. 431)) spectral sig-

nal for positive frequences only (since the waveform signal is always real). The num_frames

member of mha_spec_t (p. 465) actually denotes the number of STFT bins.

Please note that different from mha_wave_t (p. 496), a multichannel signal in mha_spec_t

(p. 465) is stored non-interleaved in the signal buffer.

Some arithmetic operations are deﬁned on struct mha_complex_t (p. 431) to facilitate efﬁcient

complex computations. The ∗=operator used here (deﬁned for real and for complex arguments)

is one of them.

3.3.4.3 Connecting the C++ class with the C plugin interface

MHAPLUGIN_CALLBACKS(example4,example4_t,spec,spec)

When connecting a class that performs spectral processing with the C interface, use spec

instead of wave as the domain indicator.

3.3 Writing openMHA Plugins. A step-by-step tutorial 21

3.3.5 example5.cpp

Many algorithms use complex operations to transform the user space variables into run time

conﬁgurations. If this takes a noticeable time (e.g. more than 100-500 µsec), the update of

the runtime conﬁguration can not take place in the real time processing thread. Furthermore,

the parallel access to complex structures may cause unpredictable results if variables are read

while only parts of them are written to memory (cf. section Consistency (p. 16)). To handle

these situations, a special C++ template class MHAPlugin::plugin_t (p. 728) was designed.

This class helps keeping all access to the conﬁguration language variables in the conﬁguration

thread rather than in the processing thread.

The runtime conﬁguration class example5_t (p. 315)is the parameter of the template class

MHAPlugin::plugin_t (p. 728). Its constructor converts the user variables into a runtime con-

ﬁguration. Because the constructor executes in the conﬁguration thread, there is no harm if the

constructor takes a long time. All other member functions and data members of the runtime

conﬁgurations are accessed only from the signal processing thread (real-time thread).

class example5_t {

public:

example5_t(unsigned int,unsigned int,mha_real_t);

mha_spec_t*process(mha_spec_t*);

private:

unsigned int channel;

mha_real_t scale;

};

The plugin interface class inherits from the plugin template class MHAPlugin::plugin_←-

t(p. 728), parameterised by the runtime conﬁguration. Conﬁguration changes (write access

to the variables) will emit a write access event of the changed variables. These events can be

connected to member functions of the interface class by the help of a MHAEvents::patchbay←-

_t (p. 512) instance.

class plugin_interface_t : public MHAPlugin::plugin_t<example5_t> {

public:

plugin_interface_t(const algo_comm_t&,const std::string&,const std::string&);

mha_spec_t*process(mha_spec_t*);

void prepare(mhaconfig_t&);

private:

void update_cfg();

/*integer variable of MHA-parser: */

MHAParser::int_t scale_ch;

/*float variable of MHA-parser: */

MHAParser::float_t factor;

/*patch bay for connecting configuration parser

events with local member functions: */

MHAEvents::patchbay_t<plugin_interface_t> patchbay;

};

The constructor of the runtime conﬁguration analyses and validates the user variables. If the

conﬁguration is invalid, an exception of type MHA_Error (p. 445) is thrown. This will cause

the openMHA conﬁguration language command which caused the change to fail: The modiﬁed

conﬁguration language variable is then reset to its original value, and the error message will

contain the message string of the MHA_Error (p. 445) exception.

22 CONTENTS

example5_t::example5_t(unsigned int ichannel,

unsigned int numchannels,

mha_real_t iscale)

: channel(ichannel),scale(iscale)

{

if( channel >= numchannels )

throw MHA_Error(__FILE__,__LINE__,

"Invalid channel number %d (only %d channels configured).",

channel,numchannels);

}

In this example, the run time conﬁguration class example5_t (p. 315)has a signal process-

ing member function. In this function, the selected channel is scaled by the given scaling factor.

mha_spec_t*example5_t::process(mha_spec_t*spec)

{

/*Scale channel number "scale_ch" by "factor": */

for(unsigned int fr = 0; fr < spec->num_frames; fr++){

spec->buf[fr + channel *spec->num_frames].re *= scale;

spec->buf[fr + channel *spec->num_frames].im *= scale;

}

return spec;

}

The constructor of the example plugin class is similar to the previous examples. A callback

triggered on write access to the variables is registered using the MHAEvents::patchbay_t

(p. 512) instance.

plugin_interface_t::plugin_interface_t(

const algo_comm_t& iac,

const std::string&,const std::string&)

: MHAPlugin::plugin_t<example5_t>("example plugin configuration structure",iac),

/*initialzing variable 'scale_ch' with MHAParser::int_t(char*name, .... ) */

scale_ch("channel number to be scaled","0","[0,["),

/*initialzing variable 'factor' with MHAParser::float_t(char*name, .... ) */

factor("scale factor","1.0","[0,2]")

{

/*Register variables to the configuration parser: */

insert_item("channel",&scale_ch);

insert_item("factor",&factor);

*On write access to the parser variables a notify callback of

*this class will be called. That funtion will update the runtime

*configuration.

patchbay.connect(&scale_ch.writeaccess,this,&plugin_interface_t::update_cfg);

patchbay.connect(&factor.writeaccess,this,&plugin_interface_t::update_cfg);

}

The processing function can gather the latest valid runtime conﬁguration by a call of poll_←-

config. On success, the class member cfg points to this conﬁguration. On error, if there is

no usable runtime conﬁguration instance, an exception is thrown. In this example, the prepare

method ensures that there is a valid runtime conﬁguration, so that in this example, no error can

be raised at this point. The prepare method is always executed before the process method is

called. The runtime conﬁguration class in this example provides a signal processing method.

The process method of the plugin interface calls the process method of this instance to perform

the actual signal processing.

3.3 Writing openMHA Plugins. A step-by-step tutorial 23

mha_spec_t*plugin_interface_t::process(mha_spec_t*spec)

{

poll_config();

return cfg->process(spec);

}

The prepare method ensures that a valid runtime conﬁguration exists by creating a new runtime

conﬁguration from the current conﬁguration language variables. If the conﬁguraion is invalid,

then an exception of type MHA_Error (p. 445) is raised and the preparation of the openMHA

fails with an error message.

void plugin_interface_t::prepare(mhaconfig_t& tfcfg)

{

if( tfcfg.domain != MHA_SPECTRUM )

throw MHA_Error(__FILE__,__LINE__,

"Example5: Only spectral processing is supported.");

/*remember the transform configuration (i.e. channel numbers): */

tftype = tfcfg;

/*make sure that a valid runtime configuration exists: */

update_cfg();

}

The update_cfg member function is called when the value of a conﬁguration language variable

changes, or from the prepare method. It allocates a new runtime conﬁguration and registers it

for later access from the real time processing thread. The function push_conﬁg (p. 726) stores

the conﬁguration in a FiFo queue of runtime conﬁgurations. Once they are inserted in the FiFo,

the MHAPlugin::plugin_t (p. 728) template is responsible for deleting runtime conﬁguration

instances stored in the FiFo. You don't need to keep track of the created instances, and you

must not delete them yourself.

void plugin_interface_t::update_cfg()

{

if( tftype.channels )

push_config(new example5_t(scale_ch.data,tftype.channels,factor.data));

}

In the end of the example code ﬁle, the macro MHAPLUGIN_CALLBACKS (p. 10) deﬁnes all

ANSI-C interface functions and passes them to the corresponding C++ class member func-

tions (partly deﬁned by the MHAPlugin::plugin_t (p. 728) template class). All exceptions of

type MHA_Error (p. 445) are caught and transformed into an appropriate error code and error

message.

MHAPLUGIN_CALLBACKS(example5,plugin_interface_t,spec,spec)

24 CONTENTS

3.3.6 example6.cpp

This last example is the same as the previous one, but it additionally creates an 'Algorithm

Communication Variable' (AC variable). It calculates the RMS level of a given channel and

stores it into this variable. The variable can be accessed by any other algorithm in the same

chain. To store the data onto disk, the 'acsave' plugin can be used. 'acmon' is a plugin which

converts AC variables into parsable monitor variables.

In the constructor of the plugin class the variable rmsdb is registered under the name

example6_rmslev as a one-dimensional AC variable of type ﬂoat. For registration of other

types, read access and other detailed informations please see Communication between al-

gorithms (p. 28).

example6_t::example6_t(const algo_comm_t& iac,

const std::string&,const std::string&)

: MHAPlugin::plugin_t<cfg_t>("example plugin configuration structure",iac),

/*initialzing variable 'channel_no' with MHAParser::int_t(char*name, .... ) */

channel_no("channel in which the RMS level is measured","0","[0,[")

{

/*Register variables to the configuration parser: */

insert_item("channel",&channel_no);

*On write access to the parser variables a notify callback of

*this class will be called. That funtion will update the runtime

*configuration.

patchbay.connect(&channel_no.writeaccess,this,&example6_t::update_cfg);

*Propagate the level variable to all algorithms in the

*processing chain. If multiple instances of this algorithm are

*required, than it is necessary to use different names for this

*variable (i.e. prefixing the name with the algorithm name

*passed to MHAInit).

ac.insert_var_float( ac.handle, "example6_rmslev", &rmsdb );

}

3.3 Writing openMHA Plugins. A step-by-step tutorial 25

3.3.7 Debugging openMHA plugins

Suppose you would want to step through the code of your openMHA plugin with a debug-

ger. This example details how to use the linux gdb debugger to inspect the example6_t←-

::prepare() (p. 317)and example6_t::process() (p. 317)routines of example6.←-

cpp (p. 24) example 6.

First, make sure that your plugin is compiled with the compiler option to include debugging

symbols: Apply the -ggdb switch to all gcc, g++ invocations.

Once the plugin is compiled, with debugging symbols, create a test conﬁguration. For example

6, assuming there is an audio ﬁle named input.wav in your working directory, you could create

a conﬁguration ﬁle named ‘debugexample6.cfg', with the following content:

# debugexample6.cfg

fragsize = 64

srate = 44100

nchannels_in = 2

iolib = MHAIOFile

io.in = input.wav

io.out = output.wav

mhalib = example6

mha.channel = 1

cmd=start

Assuming all your binaries and shared-object libraries are in your ‘bin' directory (see READ←-

ME.md), you could start gdb using

$ export MHA_LIBRARY_PATH=$PWD/bin

$ gdb $MHA_LIBRARY_PATH/mha

Set breakpoints in prepare and process methods, and start execution. Note that specifying the

breakpoint by symbol (example6_t::prepare (p. 317)) does not yet work, as the symbol

lives in the openMHA plugin that has not yet been loaded. Specifying by line number works,

however. Specifying the breakpoint by symbol also works once the plugin is loaded (i.e. when

the debugger stops in the ﬁrst break point). You can set the breakpoints like this (example

shown here is run in gdb version 7.11.1):

(gdb) run ?read:debugexample6.cfg

Starting program: {openMHA_directory}/bin/mha ?read:debugexample6.cfg

[Thread debugging using libthread_db enabled]

Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".

The Open Master Hearing Aid (openMHA) server

This program comes with ABSOLUTELY NO WARRANTY; for details see file COPYING.

This is free software, and you are welcome to redistribute it

under the terms of the GNU AFFERO GENERAL PUBLIC LICENSE, Version 3;

for details see file COPYING.

Breakpoint 1, example6_t::prepare (this=0x6478b0, tfcfg=...)

at example6.cpp:192

192 if( tfcfg.domain != MHA_WAVEFORM )

(gdb) b example6.cpp:162

Breakpoint 2 at 0x7ffff589744a: file example6.cpp, line 162.

(gdb) c

Continuing.

26 CONTENTS

Where ‘{openMHA_directory}' is the directory where openMHA is located (which should also

be your working directory in this case). Next stop is the process() method. You can now

examine and change the variables, step through the program as needed (using, for example ‘n'

to step in the next line):

Breakpoint 2, example6_t::process (this=0x7ffff6a06c0d, wave=0x10a8b550)

at example6.cpp:162

162 {

(gdb) n

163 poll_config();

(gdb)

3.4 The MHA Framework interface 27

3.4 The MHA Framework interface

28 CONTENTS

3.5 Communication between algorithms

Algorithms within one chain can share variables for communication with other algorithms.

Collaboration diagram for Communication between algorithms:

The openMHA Toolbox

library

Communication between

algorithms mha_algo_comm.h

Files

• ﬁle mha_algo_comm.h

Header ﬁle for Algorithm Communication.

Namespaces

•MHA_AC

Functions and classes for Algorithm Communication (AC) support.

Classes

• class MHA_AC::spectrum_t

Insert a MHASignal::spectrum_t (p. 800) class into the AC space.

• class MHA_AC::waveform_t

Insert a MHASignal::waveform_t (p. 812) class into the AC space.

• class MHA_AC::int_t

Insert a integer variable into the AC space.

• class MHA_AC::ﬂoat_t

Insert a ﬂoat point variable into the AC space.

• class MHA_AC::double_t

Insert a double precision ﬂoating point variable into the AC space.

• class MHA_AC::ac2matrix_t

Copy AC variable to a matrix.

• class MHA_AC::acspace2matrix_t

Copy all or a subset of all numeric AC variables into an array of matrixes.

• struct algo_comm_t

A reference handle for algorithm communication variables.

• struct comm_var_t

Algorithm communication variable structure.

3.5 Communication between algorithms 29

Functions

•mha_spec_t MHA_AC::get_var_spectrum (algo_comm_t ac, const std::string &name)

Convert an AC variable into a spectrum.

•mha_wave_t MHA_AC::get_var_waveform (algo_comm_t ac, const std::string

&name)

Convert an AC variable into a waveform.

• int MHA_AC::get_var_int (algo_comm_t ac, const std::string &name)

Return value of an integer scalar AC variable.

• ﬂoat MHA_AC::get_var_ﬂoat (algo_comm_t ac, const std::string &name)

Return value of an ﬂoating point scalar AC variable.

• std::vector<ﬂoat >MHA_AC::get_var_vﬂoat (algo_comm_t ac, const std::string

&name)

Return value of an ﬂoating point vector AC variable as standard vector of ﬂoats.

3.5.1 Detailed Description

Algorithms within one chain can share variables for communication with other algorithms.

This mechanism allows interaction between algorithms (i.e. separation of noise estimation

and noise reduction algorithms, combination of dynamic compression and noise estimation).

Through a set of simple C functions, algorithms can propagate variables of any type, even C++

classes, to other algorithms.

An algorithm communication handle (algo_comm_t (p. 201)) is passed at initialisation time to

the constructor of each plugin class constructor (p. 728). This handle contains a reference

handle, algo_comm_t::handle (p. 201), and a number of function pointers, algo_comm_t←-

::insert_var (p. 201) etc.. An algorithm communication variable is an object of type comm_←-

var_t (p. 246).

For AC variables of numeric types, openMHA Plugins for conversion into parsable monitor

variables, acmon, and storage into Matlab or text ﬁles, acsave, are available.

3.5.2 Function Documentation

3.5.2.1 mha_spec_t MHA_AC::get_var_spectrum (

algo_comm_t ac,

const std::string & name )

Convert an AC variable into a spectrum.

This function reads an AC variable and tries to convert it into a valid spectrum. The Spectrum

variable is granted to be valid only for one call of the processing function.

30 CONTENTS

Parameters

ac AC handle

name Name of the variable

Returns

Spectrum structure

3.5.2.2 mha_wave_t MHA_AC::get_var_waveform (

algo_comm_t ac,

const std::string & name )

Convert an AC variable into a waveform.

This function reads an AC variable and tries to convert it into a valid waveform. The waveform

variable is granted to be valid only for one call of the processing function.

Parameters

ac AC handle

name Name of the variable

Returns

waveform structure

3.5.2.3 int MHA_AC::get_var_int (

algo_comm_t ac,

const std::string & name )

Return value of an integer scalar AC variable.

Parameters

ac AC handle

name Name of the variable

Returns

Variable value

3.5 Communication between algorithms 31

3.5.2.4 ﬂoat MHA_AC::get_var_ﬂoat (

algo_comm_t ac,

const std::string & name )

Return value of an ﬂoating point scalar AC variable.

Parameters

ac AC handle

name Name of the variable

Returns

Variable value

3.5.2.5 std::vector<ﬂoat >MHA_AC::get_var_vﬂoat (

algo_comm_t ac,

const std::string & name )

Return value of an ﬂoating point vector AC variable as standard vector of ﬂoats.

Parameters

ac AC handle

name Name of the variable

Returns

Variable value

32 CONTENTS

3.6 Error handling in the openMHA

Errors are reported to the user via the MHA_Error (p. 445) exception.

Collaboration diagram for Error handling in the openMHA:

The openMHA Toolbox

library Error handling in the

openMHA

Classes

• class MHA_Error

Error reporting exception class.

Macros

• #deﬁne MHA_ErrorMsg(x) MHA_Error(__FILE__,__LINE__,"%s",x)

Throw an openMHA error with a text message.

• #deﬁne MHA_assert(x) if(!(x)) throw MHA_Error(__FILE__,__LINE__,"\"%s\" is

false.",#x)

Assertion macro, which throws an MHA_Error (p. 445).

• #deﬁne MHA_assert_equal(a, b) if( a != b ) throw MHA_Error(__FILE__,__LINE__←-

,"\"%s == %s\" is false (%s = %g, %s = %g).",#a,#b,#a,(double)(a),#b,(double)(b))

Equality assertion macro, which throws an MHA_Error (p. 445) with the values.

Functions

• void mha_debug (const char ∗fmt,...)

Print an info message (stderr on Linux, OutputDebugString in Windows).

3.6.1 Detailed Description

Errors are reported to the user via the MHA_Error (p. 445) exception.

3.6.2 Macro Deﬁnition Documentation

3.6.2.1 #deﬁne MHA_ErrorMsg(

x)MHA_Error(__FILE__,__LINE__,"%s",x)

Throw an openMHA error with a text message.

3.6 Error handling in the openMHA 33

Parameters

xText message.

3.6.2.2 #deﬁne MHA_assert(

x) if(!(x)) throw MHA_Error(__FILE__,__LINE__,"\"%s\" is false.",#x)

Assertion macro, which throws an MHA_Error (p. 445).

Parameters

xBoolean expression which should be true.

3.6.2.3 #deﬁne MHA_assert_equal(

b) if( a != b ) throw MHA_Error(__FILE__,__LINE__,"\"%s == %s\" is false (%s =

%g, %s = %g).",#a,#b,#a,(double)(a),#b,(double)(b))

Equality assertion macro, which throws an MHA_Error (p. 445) with the values.

Parameters

aNumeric expression which can be converted to double (for printing).

bNumeric expression which should be equal to a

3.6.3 Function Documentation

3.6.3.1 void mha_debug (

const char ∗fmt,

... )

Print an info message (stderr on Linux, OutputDebugString in Windows).

34 CONTENTS

3.7 The openMHA conﬁguration language

openMHA Plugins that should use the openMHA conﬁguration language for their conﬁguration

have to be implemented in C++ and need to include mha_parser.hh (p. 1017).

openMHA Plugins that should use the openMHA conﬁguration language for their conﬁguration

have to be implemented in C++ and need to include mha_parser.hh (p. 1017).

All required classes and functions for parser access are declared in the namespace MH←-

AParser (p. 107). The plugin class should be derived from the class MHAParser::parser_t

(p. 688) (or MHAPlugin::plugin_t (p. 728)), which symbolises a sub-parser node in the open←-

MHA script hierarchy. Variables of many types can be registered to the sub-parser node by

calling the member function insert_item (p. 690).

The openMHA Plugin template class MHAPlugin::plugin_t (p. 728) together with the Plugin

macro MHAPLUGIN_CALLBACKS (p. 10) provide the callback mappings and correct inheri-

tance. If your plugin is based on that template class, you simply have to use the insert_item

command to give access to your variables, everything else is managed internally.

A complete list of all openMHA script items is given in the description of the MHAParser (p. 107)

namespace.

3.8 The openMHA Toolbox library 35

3.8 The openMHA Toolbox library

The openMHA toolbox is a static C++ library which makes it more comfortable to develop

openMHA plugins.

Collaboration diagram for The openMHA Toolbox library:

Vector and matrix processing

toolbox

Complex arithmetics

in the openMHA

The openMHA Toolbox

library

MHASignal

Error handling in the

openMHA

Fast Fourier Transform

functions

Communication between

algorithms

mha_algo_comm.h

Modules

•Error handling in the openMHA

Errors are reported to the user via the MHA_Error (p. 445) exception.

•Vector and matrix processing toolbox

The vector and matrix processing toolbox consists of a number of classes deﬁned in the

namespace MHASignal (p. 118), and many functions and operators for use with the struc-

tures mha_wave_t (p. 496) and mha_spec_t (p. 465).

•Complex arithmetics in the openMHA

•Fast Fourier Transform functions

Files

• ﬁle mha_algo_comm.h

Header ﬁle for Algorithm Communication.

• ﬁle mha_ﬁlter.hh

Header ﬁle for IIR ﬁlter classes.

• ﬁle mha_signal.hh

Header ﬁle for audio signal handling and processing classes.

• ﬁle mha_tablelookup.hh

Header ﬁle for table lookup classes.

36 CONTENTS

Namespaces

•MHAOvlFilter

Namespace for overlapping FFT based ﬁlter bank classes and functions.

•MHAFilter

Namespace for IIR and FIR ﬁlter classes.

•MHAParser

Name space for the openMHA-Parser conﬁguration language.

•MHASignal

Namespace for audio signal handling and processing classes.

•MHATableLookup

Namespace for table lookup classes.

3.8.1 Detailed Description

The openMHA toolbox is a static C++ library which makes it more comfortable to develop

openMHA plugins.

It contains the openMHA script language classes.

3.9 Vector and matrix processing toolbox 37

3.9 Vector and matrix processing toolbox

The vector and matrix processing toolbox consists of a number of classes deﬁned in the names-

pace MHASignal (p. 118), and many functions and operators for use with the structures mha←-

_wave_t (p. 496) and mha_spec_t (p. 465).

Collaboration diagram for Vector and matrix processing toolbox:

Vector and matrix processing

toolbox The openMHA Toolbox

library

MHASignal

Namespaces

•MHASignal

Namespace for audio signal handling and processing classes.

•MHAWindow

Collection of Window types.

Classes

• struct mha_wave_t

Waveform signal structure.

• struct mha_spec_t

Spectrum signal structure.

• struct mha_audio_descriptor_t

Description of an audio fragment (planned as a replacement of mhaconﬁg_t (p. 504)).

• struct mha_audio_t

An audio fragment in the openMHA (planned as a replacement of mha_wave_t (p. 496) and

mha_spec_t (p. 465)).

• class MHASignal::spectrum_t

a signal processing class for spectral data (based on mha_spec_t (p. 465))

• class MHASignal::waveform_t

signal processing class for waveform data (based on mha_wave_t (p. 496))

• class MHASignal::doublebuffer_t

Double-buffering class.

• class MHASignal::hilbert_t

Hilbert transformation of a waveform segment.

• class MHASignal::minphase_t

Minimal phase function.

• class MHASignal::uint_vector_t

38 CONTENTS

Vector of unsigned values, used for size and index description of n-dimensional matrixes.

• class MHASignal::matrix_t

n-dimensional matrix with real or complex ﬂoating point values.

• class MHAParser::window_t

MHA conﬁguration interface for a window function generator.

• class MHASignal::delay_wave_t

Delayline containing wave fragments.

• class MHASignal::async_rmslevel_t

Class for asynchronous level metering.

Typedefs

• typedef ﬂoat mha_real_t

openMHA type for real numbers

Functions

•mha_wave_t range (mha_wave_t s, unsigned int k0, unsigned int len)

Return a time interval from a waveform chunk.

•mha_spec_t channels (mha_spec_t s, unsigned int ch_start, unsigned int nch)

Return a channel interval from a spectrum.

• void MHASignal::for_each (mha_wave_t ∗s, mha_real_t(∗fun)(mha_real_t))

Apply a function to each element of a mha_wave_t (p. 496).

•mha_real_t MHASignal::lin2db (mha_real_t x)

Conversion from linear scale to dB (no SPL reference)

•mha_real_t MHASignal::db2lin (mha_real_t x)

Conversion from dB scale to linear (no SPL reference)

•mha_real_t MHASignal::pa2dbspl (mha_real_t x)

Conversion from linear Pascal scale to dB SPL.

•mha_real_t MHASignal::pa22dbspl (mha_real_t x, mha_real_t eps=1e-20f)

Conversion from squared Pascal scale to dB SPL.

•mha_real_t MHASignal::dbspl2pa (mha_real_t x)

Conversion from dB SPL to linear Pascal scale.

•mha_real_t MHASignal::smp2sec (mha_real_t n, mha_real_t srate)

conversion from samples to seconds

•mha_real_t MHASignal::sec2smp (mha_real_t sec, mha_real_t srate)

conversion from seconds to samples

•mha_real_t MHASignal::bin2freq (mha_real_t bin, unsigned fftlen, mha_real_t srate)

conversion from fft bin index to frequency

•mha_real_t MHASignal::freq2bin (mha_real_t freq, unsigned fftlen, mha_real_t srate)

conversion from frequency to fft bin index

•mha_real_t MHASignal::smp2rad (mha_real_t samples, unsigned bin, unsigned fftlen)

conversion from delay in samples to phase shift

3.9 Vector and matrix processing toolbox 39

•mha_real_t MHASignal::rad2smp (mha_real_t phase_shift, unsigned bin, unsigned

fftlen)

conversion from phase shift to delay in samples

•template<class elem_type >

std::vector<elem_type >MHASignal::dupvec (std::vector<elem_type >vec, unsigned

Duplicate last vector element to match desired size.

•template<class elem_type >

std::vector<elem_type >MHASignal::dupvec_chk (std::vector<elem_type >vec, un-

signed n)

Duplicate last vector element to match desired size, check for dimension.

• bool equal_dim (const mha_wave_t &a, const mha_wave_t &b)

Test for equal dimension of waveform structures.

• bool equal_dim (const mha_wave_t &a, const mhaconﬁg_t &b)

Test for match of waveform dimension with mhaconﬁg structure.

• bool equal_dim (const mha_spec_t &a, const mha_spec_t &b)

Test for equal dimension of spectrum structures.

• bool equal_dim (const mha_spec_t &a, const mhaconﬁg_t &b)

Test for match of spectrum dimension with mhaconﬁg structure.

• bool equal_dim (const mha_wave_t &a, const mha_spec_t &b)

Test for equal dimension of waveform/spectrum structures.

• bool equal_dim (const mha_spec_t &a, const mha_wave_t &b)

Test for equal dimension of waveform/spectrum structures.

• void integrate (mha_wave_t &s)

Numeric integration of a signal vector (real values)

• void integrate (mha_spec_t &s)

Numeric integration of a signal vector (complex values)

• unsigned int size (const mha_wave_t &s)

Return size of a waveform structure.

• unsigned int size (const mha_spec_t &s)

Return size of a spectrum structure.

• unsigned int size (const mha_wave_t ∗s)

Return size of a waveform structure.

• unsigned int size (const mha_spec_t ∗s)

Return size of a spectrum structure.

• void clear (mha_wave_t &s)

Set all values of waveform to zero.

• void clear (mha_wave_t ∗s)

Set all values of waveform to zero.

• void clear (mha_spec_t &s)

Set all values of spectrum to zero.

• void clear (mha_spec_t ∗s)

Set all values of spectrum to zero.

• void assign (mha_wave_t self, mha_real_t val)

Set all values of waveform 'self' to 'val'.

40 CONTENTS

• void assign (mha_wave_t self, const mha_wave_t &val)

Set all values of waveform 'self' to 'val'.

• void assign (mha_spec_t self, const mha_spec_t &val)

Set all values of spectrum 'self' to 'val'.

• void timeshift (mha_wave_t &self, int shift)

Time shift of waveform chunk.

•mha_real_t &value (mha_wave_t ∗s, unsigned int fr, unsigned int ch)

Access an element of a waveform structure.

• const mha_real_t &value (const mha_wave_t ∗s, unsigned int fr, unsigned int ch)

Constant access to an element of a waveform structure.

•mha_complex_t &value (mha_spec_t ∗s, unsigned int fr, unsigned int ch)

Access to an element of a spectrum.

• const mha_complex_t &value (const mha_spec_t ∗s, unsigned int fr, unsigned int ch)

Constant access to an element of a spectrum.

•mha_real_t &value (mha_wave_t &s, unsigned int fr, unsigned int ch)

Access to an element of a waveform structure.

• const mha_real_t &value (const mha_wave_t &s, unsigned int fr, unsigned int ch)

Constant access to an element of a waveform structure.

•mha_complex_t &value (mha_spec_t &s, unsigned int fr, unsigned int ch)

Access to an element of a spectrum.

• const mha_complex_t &value (const mha_spec_t &s, unsigned int fr, unsigned int ch)

Constant access to an element of a spectrum.

• std::vector<ﬂoat >std_vector_ﬂoat (const mha_wave_t &)

Converts a mha_wave_t (p. 496) structure into a std::vector<ﬂoat>(interleaved order).

• std::vector<std::vector<ﬂoat > > std_vector_vector_ﬂoat (const mha_wave_t &)

Converts a mha_wave_t (p. 496) structure into a std::vector<std::vector<ﬂoat> > (outer vec-

tor represents channels).

• std::vector<std::vector<mha_complex_t > > std_vector_vector_complex (const

mha_spec_t &)

Converts a mha_spec_t (p. 465) structure into a std::vector<std::vector<mha_complex_t> >

(outer vector represents channels).

•mha_wave_t &operator+= (mha_wave_t &, const mha_real_t &)

Addition operator.

•mha_wave_t &operator+= (mha_wave_t &, const mha_wave_t &)

Addition operator.

•mha_wave_t &operator-= (mha_wave_t &, const mha_wave_t &)

Subtraction operator.

•mha_spec_t &operator-= (mha_spec_t &, const mha_spec_t &)

Subtraction operator.

•mha_wave_t &operator∗=(mha_wave_t &, const mha_real_t &)

Element-wise multiplication operator.

•mha_wave_t &operator∗=(mha_wave_t &, const mha_wave_t &)

Element-wise multiplication operator.

•mha_spec_t &operator∗=(mha_spec_t &, const mha_real_t &)

Element-wise multiplication operator.

3.9 Vector and matrix processing toolbox 41

•mha_spec_t &operator∗=(mha_spec_t &, const mha_wave_t &)

Element-wise multiplication operator.

•mha_spec_t &operator∗=(mha_spec_t &, const mha_spec_t &)

Element-wise multiplication operator.

•mha_spec_t &operator/= (mha_spec_t &, const mha_spec_t &)

Element-wise division operator.

•mha_wave_t &operator/= (mha_wave_t &, const mha_wave_t &)

Element-wise division operator.

•mha_spec_t &operator+= (mha_spec_t &, const mha_spec_t &)

Addition operator.

•mha_spec_t &operator+= (mha_spec_t &, const mha_real_t &)

Addition operator.

•mha_wave_t &operator∧=(mha_wave_t &self, const mha_real_t &arg)

Exponent operator.

• void MHASignal::copy_channel (mha_spec_t &self, const mha_spec_t &src, unsigned

sch, unsigned dch)

Copy one channel of a source signal.

• void MHASignal::copy_channel (mha_wave_t &self, const mha_wave_t &src, un-

signed src_channel, unsigned dest_channel)

Copy one channel of a source signal.

•mha_real_t MHASignal::rmslevel (const mha_spec_t &s, unsigned int channel, un-

signed int fftlen)

Return RMS level of a spectrum channel.

•mha_real_t MHASignal::colored_intensity (const mha_spec_t &s, unsigned int chan-

nel, unsigned int fftlen, mha_real_t sqfreq_response[ ])

Colored spectrum intensity.

•mha_real_t MHASignal::maxabs (const mha_spec_t &s, unsigned int channel)

Find maximal absolute value.

•mha_real_t MHASignal::rmslevel (const mha_wave_t &s, unsigned int channel)

Return RMS level of a waveform channel.

•mha_real_t MHASignal::maxabs (const mha_wave_t &s, unsigned int channel)

Find maximal absolute value.

•mha_real_t MHASignal::maxabs (const mha_wave_t &s)

Find maximal absolute value.

•mha_real_t MHASignal::max (const mha_wave_t &s)

Find maximal value.

•mha_real_t MHASignal::min (const mha_wave_t &s)

Find minimal value.

•mha_real_t MHASignal::sumsqr_channel (const mha_wave_t &s, unsigned int chan-

nel)

Calculate sum of squared values in one channel.

•mha_real_t MHASignal::sumsqr_frame (const mha_wave_t &s, unsigned int frame)

Calculate sum over all channels of squared values.

• void conjugate (mha_spec_t &self)

Replace (!) the value of this mha_spec_t (p. 465) with its conjugate.

42 CONTENTS

3.9.1 Detailed Description

The vector and matrix processing toolbox consists of a number of classes deﬁned in the names-

pace MHASignal (p. 118), and many functions and operators for use with the structures mha←-

_wave_t (p. 496) and mha_spec_t (p. 465).

3.9.2 Typedef Documentation

3.9.2.1 typedef ﬂoat mha_real_t

openMHA type for real numbers

This type is expected to be allways the C-type 'ﬂoat' (IEEE 754 single).

3.9.3 Function Documentation

3.9.3.1 mha_wave_t range (

mha_wave_t s,

unsigned int k0,

unsigned int len )

Return a time interval from a waveform chunk.

A waveform chunk containing a time intervall of a larger waveform chunk is returned. The

number of channels remains constant. The data of the output waveform structure points to

the data of the input structure, i.e., write access to the output waveform chunk modiﬁes the

corresponding entries in the input chunk.

Parameters

sWaveform structure

k0 Index of ﬁrst value in output

len Number of frames in output

Returns

Waveform structure representing the sub-interval.

3.9.3.2 mha_spec_t channels (

mha_spec_t s,

unsigned int ch_start,

unsigned int nch )

Return a channel interval from a spectrum.

3.9 Vector and matrix processing toolbox 43

Parameters

sInput spectrum

ch_start Index of ﬁrst channel in output

nch Number of channels in output

Returns

Spectrum structure representing the sub-interval.

3.9.3.3 void MHASignal::for_each (

mha_wave_t ∗s,

mha_real_t(∗)(mha_real_t)fun )[inline]

Apply a function to each element of a mha_wave_t (p. 496).

Parameters

sPointer to a mha_wave_t (p. 496) structure

fun Function to be applied (one argument)

3.9.3.4 mha_real_t MHASignal::lin2db (

mha_real_t x)[inline]

Conversion from linear scale to dB (no SPL reference)

Parameters

xLinear input.

3.9.3.5 mha_real_t MHASignal::db2lin (

mha_real_t x)[inline]

Conversion from dB scale to linear (no SPL reference)

Parameters

xdB input.

3.9.3.6 mha_real_t MHASignal::pa2dbspl (

mha_real_t x)[inline]

Conversion from linear Pascal scale to dB SPL.

44 CONTENTS

Parameters

xLinear input.

3.9.3.7 mha_real_t MHASignal::pa22dbspl (

mha_real_t x,

mha_real_t eps = 1e-20f )[inline]

Conversion from squared Pascal scale to dB SPL.

Parameters

xsquared pascal input

eps minimum squared-pascal value

3.9.3.8 mha_real_t MHASignal::dbspl2pa (

mha_real_t x)[inline]

Conversion from dB SPL to linear Pascal scale.

Parameters

xLinear input.

3.9.3.9 mha_real_t MHASignal::smp2sec (

mha_real_t n,

mha_real_t srate )[inline]

conversion from samples to seconds

Parameters

nnumber of samples

srate sampling rate / Hz

3.9.3.10 mha_real_t MHASignal::sec2smp (

mha_real_t sec,

mha_real_t srate )[inline]

conversion from seconds to samples

Parameters

sec time in seconds

srate sampling rate / Hz

3.9 Vector and matrix processing toolbox 45

Returns

number of samples, generally has non-zero fractional part

3.9.3.11 mha_real_t MHASignal::bin2freq (

mha_real_t bin,

unsigned fftlen,

mha_real_t srate )[inline]

conversion from fft bin index to frequency

Parameters

bin index of fft bin, index 0 has dc

fftlen FFT length

srate sampling frequency / Hz

Returns

frequency of fft bin / Hz

3.9.3.12 mha_real_t MHASignal::freq2bin (

mha_real_t freq,

unsigned fftlen,

mha_real_t srate )[inline]

conversion from frequency to fft bin index

Parameters

freq frequency / Hz

fftlen FFT length

srate sampling frequency / Hz

Returns

0-based index of fft bin, generally has non-zero fractional part

3.9.3.13 mha_real_t MHASignal::smp2rad (

mha_real_t samples,

unsigned bin,

unsigned fftlen )[inline]

conversion from delay in samples to phase shift

Compute phase shift that needs to be applied to fft spectrum to achieve the desired delay.

46 CONTENTS

Parameters

samples delay in samples. Positive delay: shift current signal to future.

bin index of fft bin, index 0 has dc (index 0 and nyqvist bin cannot be delayed)

fftlen FFT length

Returns

The phase shift in radiant that needs to be applied to fft bin to achieve the desired delay. A

positive delay requires a negative phase shift. If required phase shift is >pi or <-pi, then

the desired delay cannot be applied in the fft domain with given parameters. Required

phase shifts close to pi should not be used. If bin is 0 or nyqvist, returns 0 phase shift.

3.9.3.14 mha_real_t MHASignal::rad2smp (

mha_real_t phase_shift,

unsigned bin,

unsigned fftlen )[inline]

conversion from phase shift to delay in samples

Compute delay in samples that is achieved by a phase shift.

Parameters

phase_shift phase shift in radiant

bin index of fft bin, index 0 has dc (index 0 and nyqvist bin cannot be delayed)

fftlen FFT length

Returns

The delay in samples achieved by applying the phase shift. A negative phase shift causes

a positive delay: shifts current signal to future.

3.9.3.15 template<class elem_type >std::vector<elem_type>MHASignal::dupvec (

std::vector<elem_type >vec,

unsigned n)

Duplicate last vector element to match desired size.

Parameters

vec Input vector.

nTarget number of elements.

3.9 Vector and matrix processing toolbox 47

Return values

Resized vector.

3.9.3.16 template<class elem_type >std::vector<elem_type>MHASignal::dupvec_chk (

std::vector<elem_type >vec,

unsigned n)

Duplicate last vector element to match desired size, check for dimension.

The input dimension can be either 1 or the target length.

Parameters

vec Input vector.

nTarget number of elements.

Return values

Resized vector.

3.9.3.17 bool equal_dim (

const mha_wave_t &a,

const mha_wave_t &b)[inline]

Test for equal dimension of waveform structures.

3.9.3.18 bool equal_dim (

const mha_wave_t &a,

const mhaconﬁg_t &b)[inline]

Test for match of waveform dimension with mhaconﬁg structure.

3.9.3.19 bool equal_dim (

const mha_spec_t &a,

const mha_spec_t &b)[inline]

Test for equal dimension of spectrum structures.

3.9.3.20 bool equal_dim (

const mha_spec_t &a,

const mhaconﬁg_t &b)[inline]

Test for match of spectrum dimension with mhaconﬁg structure.

48 CONTENTS

3.9.3.21 bool equal_dim (

const mha_wave_t &a,

const mha_spec_t &b)[inline]

Test for equal dimension of waveform/spectrum structures.

Warning

Waveform structures mha_wave_t (p. 496) use interleaved data order, while spectrum

structures mha_spec_t (p. 465) use non-interleaved.

3.9.3.22 bool equal_dim (

const mha_spec_t &a,

const mha_wave_t &b)[inline]

Test for equal dimension of waveform/spectrum structures.

Warning

Waveform structures mha_wave_t (p. 496) use interleaved data order, while spectrum

structures mha_spec_t (p. 465) use non-interleaved.

3.9.3.23 void integrate (

mha_wave_t &s)

Numeric integration of a signal vector (real values)

Parameters

sInput signal vector

3.9.3.24 void integrate (

mha_spec_t &s)

Numeric integration of a signal vector (complex values)

Parameters

sInput signal vector

3.9.3.25 unsigned int size (

const mha_wave_t &s)[inline]

Return size of a waveform structure.

3.9 Vector and matrix processing toolbox 49

3.9.3.26 unsigned int size (

const mha_spec_t &s)[inline]

Return size of a spectrum structure.

3.9.3.27 unsigned int size (

const mha_wave_t ∗s)[inline]

Return size of a waveform structure.

3.9.3.28 unsigned int size (

const mha_spec_t ∗s)[inline]

Return size of a spectrum structure.

3.9.3.29 void clear (

mha_wave_t &s)[inline]

Set all values of waveform to zero.

3.9.3.30 void clear (

mha_wave_t ∗s)[inline]

Set all values of waveform to zero.

3.9.3.31 void clear (

mha_spec_t &s)[inline]

Set all values of spectrum to zero.

3.9.3.32 void clear (

mha_spec_t ∗s)[inline]

Set all values of spectrum to zero.

3.9.3.33 void assign (

mha_wave_t self,

mha_real_t val )[inline]

Set all values of waveform 'self' to 'val'.

Parameters

self Waveform to be modiﬁed.

val Value to be assigned to all entries of waveform.

50 CONTENTS

3.9.3.34 void assign (

mha_wave_t self,

const mha_wave_t &val )

Set all values of waveform 'self' to 'val'.

Parameters

self Waveform to be modiﬁed.

val Source waveform structure.

3.9.3.35 void assign (

mha_spec_t self,

const mha_spec_t &val )

Set all values of spectrum 'self' to 'val'.

Parameters

self Spectrum to be modiﬁed.

val Source spectrum.

3.9.3.36 void timeshift (

mha_wave_t &self,

int shift )

Time shift of waveform chunk.

Shifted areas are ﬁlled with zeros.

Parameters

self Waveform chunk to be shifted

shift Shift amount, positive values shift to later times

3.9.3.37 mha_real_t& value (

mha_wave_t ∗s,

unsigned int fr,

unsigned int ch )[inline]

Access an element of a waveform structure.

Parameters

sWaveform structure

fr Frame number

ch Channel number

3.9 Vector and matrix processing toolbox 51

Returns

Reference to element

3.9.3.38 const mha_real_t& value (

const mha_wave_t ∗s,

unsigned int fr,

unsigned int ch )[inline]

Constant access to an element of a waveform structure.

Parameters

sWaveform structure

fr Frame number

ch Channel number

Returns

Reference to element

3.9.3.39 mha_complex_t& value (

mha_spec_t ∗s,

unsigned int fr,

unsigned int ch )[inline]

Access to an element of a spectrum.

Parameters

sSpectrum structure

fr Bin number

ch Channel number

Returns

Reference to element

3.9.3.40 const mha_complex_t& value (

const mha_spec_t ∗s,

unsigned int fr,

unsigned int ch )[inline]

Constant access to an element of a spectrum.

52 CONTENTS

Parameters

sSpectrum structure

fr Bin number

ch Channel number

Returns

Reference to element

3.9.3.41 mha_real_t& value (

mha_wave_t &s,

unsigned int fr,

unsigned int ch )[inline]

Access to an element of a waveform structure.

Parameters

sWaveform structure

fr Frame number

ch Channel number

Returns

Reference to element

3.9.3.42 const mha_real_t& value (

const mha_wave_t &s,

unsigned int fr,

unsigned int ch )[inline]

Constant access to an element of a waveform structure.

Parameters

sWaveform structure

fr Frame number

ch Channel number

Returns

Reference to element

3.9 Vector and matrix processing toolbox 53

3.9.3.43 mha_complex_t& value (

mha_spec_t &s,

unsigned int fr,

unsigned int ch )[inline]

Access to an element of a spectrum.

Parameters

sSpectrum structure

fr Bin number

ch Channel number

Returns

Reference to element

3.9.3.44 const mha_complex_t& value (

const mha_spec_t &s,

unsigned int fr,

unsigned int ch )[inline]

Constant access to an element of a spectrum.

Parameters

sSpectrum structure

fr Bin number

ch Channel number

Returns

Reference to element

3.9.3.45 std::vector<ﬂoat>std_vector_ﬂoat (

const mha_wave_t & )

Converts a mha_wave_t (p. 496) structure into a std::vector<ﬂoat>(interleaved order).

Warning

This function is not real-time safe. Do not use in signal processing thread.

54 CONTENTS

3.9.3.46 std::vector<std::vector<ﬂoat> > std_vector_vector_ﬂoat (

const mha_wave_t & )

Converts a mha_wave_t (p. 496) structure into a std::vector<std::vector<ﬂoat> > (outer vec-

tor represents channels).

Warning

This function is not real-time safe. Do not use in signal processing thread.

3.9.3.47 std::vector<std::vector<mha_complex_t> > std_vector_vector_complex (

const mha_spec_t & )

Converts a mha_spec_t (p. 465) structure into a std::vector<std::vector<mha_complex_t> >

(outer vector represents channels).

Warning

This function is not real-time safe. Do not use in signal processing thread.

3.9.3.48 mha_wave_t& operator+= (

mha_wave_t &,

const mha_real_t & )

Addition operator.

3.9.3.49 mha_wave_t& operator+= (

mha_wave_t &,

const mha_wave_t & )

Addition operator.

3.9.3.50 mha_wave_t& operator-= (

mha_wave_t &,

const mha_wave_t & )

Subtraction operator.

3.9.3.51 mha_spec_t& operator-= (

mha_spec_t &,

const mha_spec_t & )

Subtraction operator.

3.9 Vector and matrix processing toolbox 55

3.9.3.52 mha_wave_t& operator∗= (

mha_wave_t &,

const mha_real_t & )

Element-wise multiplication operator.

3.9.3.53 mha_wave_t& operator∗= (

mha_wave_t &,

const mha_wave_t & )

Element-wise multiplication operator.

3.9.3.54 mha_spec_t& operator∗= (

mha_spec_t &,

const mha_real_t & )

Element-wise multiplication operator.

3.9.3.55 mha_spec_t& operator∗= (

mha_spec_t &,

const mha_wave_t & )

Element-wise multiplication operator.

3.9.3.56 mha_spec_t& operator∗= (

mha_spec_t &,

const mha_spec_t & )

Element-wise multiplication operator.

3.9.3.57 mha_spec_t& operator/= (

mha_spec_t &,

const mha_spec_t & )

Element-wise division operator.

3.9.3.58 mha_wave_t& operator/= (

mha_wave_t &,

const mha_wave_t & )

Element-wise division operator.

3.9.3.59 mha_spec_t& operator+= (

mha_spec_t &,

const mha_spec_t & )

Addition operator.

56 CONTENTS

3.9.3.60 mha_spec_t& operator+= (

mha_spec_t &,

const mha_real_t & )

Addition operator.

3.9.3.61 mha_wave_t& operator∧= (

mha_wave_t &self,

const mha_real_t &arg )

Exponent operator.

Warning

This overwrites the xor operator!

3.9.3.62 void MHASignal::copy_channel (

mha_spec_t &self,

const mha_spec_t &src,

unsigned sch,

unsigned dch )

Copy one channel of a source signal.

Parameters

self Destination.

src Source

sch Source channel number

dch Destination channel number

3.9.3.63 void MHASignal::copy_channel (

mha_wave_t &self,

const mha_wave_t &src,

unsigned src_channel,

unsigned dest_channel )

Copy one channel of a source signal.

Parameters

self Destination.

src Source

src_channel Source channel number

dest_channel Destination channel number

3.9 Vector and matrix processing toolbox 57

3.9.3.64 mha_real_t MHASignal::rmslevel (

const mha_spec_t &s,

unsigned int channel,

unsigned int fftlen )

Return RMS level of a spectrum channel.

Parameters

sInput spectrum

channel Channel number to be tested

fftlen FFT length (to correctly count the level of the Nyquist bin)

Returns

RMS level in Pa

3.9.3.65 mha_real_t MHASignal::colored_intensity (

const mha_spec_t &s,

unsigned int channel,

unsigned int fftlen,

mha_real_t sqfreq_response[ ] )

Colored spectrum intensity.

computes the squared sum of the spectrum after ﬁltering with the frequency response

Parameters

sInput spectrum

channel Channel number to be tested

fftlen FFT length (to correctly count the level of the Nyquist bin)

sqfreq_response A squared weighting factor for every fft bin.

Returns

sum of squares. Root of this is the colored level in Pa

3.9.3.66 mha_real_t MHASignal::maxabs (

const mha_spec_t &s,

unsigned int channel )

Find maximal absolute value.

58 CONTENTS

Parameters

sInput signal

channel Channel to be tested

Returns

maximum absolute value

3.9.3.67 mha_real_t MHASignal::rmslevel (

const mha_wave_t &s,

unsigned int channel )

Return RMS level of a waveform channel.

Parameters

sInput waveform signal

channel Channel number to be tested

Returns

RMS level in Pa

3.9.3.68 mha_real_t MHASignal::maxabs (

const mha_wave_t &s,

unsigned int channel )

Find maximal absolute value.

Parameters

sInput signal

channel Channel to be tested

Returns

maximum absolute value

3.9.3.69 mha_real_t MHASignal::maxabs (

const mha_wave_t &s)

Find maximal absolute value.

3.9 Vector and matrix processing toolbox 59

Parameters

sInput signal

Returns

maximum absolute value

3.9.3.70 mha_real_t MHASignal::max (

const mha_wave_t &s)

Find maximal value.

Parameters

sInput signal

Returns

maximum absolute value

3.9.3.71 mha_real_t MHASignal::min (

const mha_wave_t &s)

Find minimal value.

Parameters

sInput signal

Returns

maximum absolute value

3.9.3.72 mha_real_t MHASignal::sumsqr_channel (

const mha_wave_t &s,

unsigned int channel )

Calculate sum of squared values in one channel.

Parameters

sInput signal

channel Channel

60 CONTENTS

Returns

Px2

3.9.3.73 mha_real_t MHASignal::sumsqr_frame (

const mha_wave_t &s,

unsigned int frame )

Calculate sum over all channels of squared values.

Parameters

sInput signal

frame Frame number

Returns

Px2

3.9.3.74 void conjugate (

mha_spec_t &self )[inline]

Replace (!) the value of this mha_spec_t (p. 465) with its conjugate.

3.10 Complex arithmetics in the openMHA 61

3.10 Complex arithmetics in the openMHA

Collaboration diagram for Complex arithmetics in the openMHA:

Complex arithmetics

in the openMHA

The openMHA Toolbox

library

Classes

• struct mha_complex_t

Type for complex ﬂoating point values.

Functions

•mha_complex_t &set (mha_complex_t &self, mha_real_t real, mha_real_t imag=0)

Assign real and imaginary parts to a mha_complex_t (p. 431) variable.

•mha_complex_t mha_complex (mha_real_t real, mha_real_t imag=0)

Create a new mha_complex_t (p. 431) with speciﬁed real and imaginary parts.

•mha_complex_t &set (mha_complex_t &self, const std::complex<mha_real_t >

&stdcomplex)

Assign a mha_complex_t (p. 431) variable from a std::complex.

• std::complex<mha_real_t >stdcomplex (const mha_complex_t &self)

Create a std::complex from mha_complex_t (p. 431).

•mha_complex_t &expi (mha_complex_t &self, mha_real_t angle)

replaces the value of the given mha_complex_t (p. 431) with exp(i∗b).

• double angle (const mha_complex_t &self)

Computes the angle of a complex number in the complex plane.

•mha_complex_t &operator+= (mha_complex_t &self, const mha_complex_t &other)

Addition of two complex numbers, overwriting the ﬁrst.

•mha_complex_t operator+ (const mha_complex_t &self, const mha_complex_←-

t&other)

Addition of two complex numbers, result is a temporary object.

•mha_complex_t &operator+= (mha_complex_t &self, mha_real_t other_real)

Addition of a complex and a real number, overwriting the complex.

•mha_complex_t operator+ (const mha_complex_t &self, mha_real_t other_real)

Addition of a complex and a real number, result is a temporary object.

•mha_complex_t &operator-= (mha_complex_t &self, const mha_complex_t &other)

Subtraction of two complex numbers, overwriting the ﬁrst.

62 CONTENTS

•mha_complex_t operator- (const mha_complex_t &self, const mha_complex_←-

t&other)

Subtraction of two complex numbers, result is a temporary object.

•mha_complex_t &operator-= (mha_complex_t &self, mha_real_t other_real)

Subtraction of a complex and a real number, overwriting the complex.

•mha_complex_t operator- (const mha_complex_t &self, mha_real_t other_real)

Subtraction of a complex and a real number, result is a temporary object.

•mha_complex_t &operator∗=(mha_complex_t &self, const mha_complex_t &other)

Multiplication of two complex numbers, overwriting the ﬁrst.

•mha_complex_t operator∗(const mha_complex_t &self, const mha_complex_←-

t&other)

Multiplication of two complex numbers, result is a temporary object.

•mha_complex_t &operator∗=(mha_complex_t &self, mha_real_t other_real)

Multiplication of a complex and a real number, overwriting the complex.

•mha_complex_t &expi (mha_complex_t &self, mha_real_t angle,mha_real_t factor)

replaces (!) the value of the given mha_complex_t (p. 431) with a ∗exp(i∗b)

•mha_complex_t operator∗(const mha_complex_t &self, mha_real_t other_real)

Multiplication of a complex and a real number, result is a temporary object.

•mha_real_t abs2 (const mha_complex_t &self)

Compute the square of the absolute value of a complex value.

•mha_real_t abs (const mha_complex_t &self)

Compute the absolute value of a complex value.

•mha_complex_t &operator/= (mha_complex_t &self, mha_real_t other_real)

Division of a complex and a real number, overwriting the complex.

•mha_complex_t operator/ (const mha_complex_t &self, mha_real_t other_real)

Division of a complex and a real number, result is a temporary object.

•mha_complex_t &safe_div (mha_complex_t &self, const mha_complex_t &other,

mha_real_t eps, mha_real_t eps2)

Safe division of two complex numbers, overwriting the ﬁrst.

•mha_complex_t &operator/= (mha_complex_t &self, const mha_complex_t &other)

Division of two complex numbers, overwriting the ﬁrst.

•mha_complex_t operator/ (const mha_complex_t &self, const mha_complex_←-

t&other)

Division of two complex numbers, result is a temporary object.

•mha_complex_t operator- (const mha_complex_t &self)

Unary minus on a complex results in a negative temporary object.

• bool operator== (const mha_complex_t &x, const mha_complex_t &y)

Compare two complex numbers for equality.

• bool operator!= (const mha_complex_t &x, const mha_complex_t &y)

Compare two complex numbers for inequality.

• void conjugate (mha_complex_t &self)

Replace (!) the value of this mha_complex_t (p. 431) with its conjugate.

•mha_complex_t _conjugate (const mha_complex_t &self)

Compute the cojugate of this complex value.

• void reciprocal (mha_complex_t &self)

3.10 Complex arithmetics in the openMHA 63

Replace the value of this complex with its reciprocal.

•mha_complex_t _reciprocal (const mha_complex_t &self)

compute the reciprocal of this complex value.

• void normalize (mha_complex_t &self)

Divide a complex by its absolute value, thereby normalizing it (projecting onto the unit circle).

• void normalize (mha_complex_t &self, mha_real_t margin)

Divide a complex by its absolute value, thereby normalizing it (projecting onto the unit circle),

with a safety margin.

• bool almost (const mha_complex_t &self, const mha_complex_t &other, mha_real_t

times_epsilon=1e2)

Compare two complex numbers for equality except for a small relative error.

• bool operator<(const mha_complex_t &x, const mha_complex_t &y)

Compares the absolute values of two complex numbers.

3.10.1 Detailed Description

3.10.2 Function Documentation

3.10.2.1 mha_complex_t& set (

mha_complex_t &self,

mha_real_t real,

mha_real_t imag = 0)[inline]

Assign real and imaginary parts to a mha_complex_t (p. 431) variable.

Parameters

self The mha_complex_t (p. 431) variable whose value is about to change.

real The new real part.

imag The new imaginary part.

Returns

A reference to the changed variable.

3.10.2.2 mha_complex_t mha_complex (

mha_real_t real,

mha_real_t imag = 0)[inline]

Create a new mha_complex_t (p. 431) with speciﬁed real and imaginary parts.

Parameters

real The real part.

imag The imaginary part.

64 CONTENTS

Returns

The new value.

3.10.2.3 mha_complex_t& set (

mha_complex_t &self,

const std::complex<mha_real_t >&stdcomplex )[inline]

Assign a mha_complex_t (p. 431) variable from a std::complex.

Parameters

self The mha_complex_t (p. 431) variable whose value is about to change.

stdcomplex The new complex value.

Returns

A reference to the changed variable.

3.10.2.4 std::complex<mha_real_t>stdcomplex (

const mha_complex_t &self )[inline]

Create a std::complex from mha_complex_t (p. 431).

3.10.2.5 mha_complex_t& expi (

mha_complex_t &self,

mha_real_t angle )[inline]

replaces the value of the given mha_complex_t (p. 431) with exp(i∗b).

Parameters

self The mha_complex_t (p. 431) variable whose value is about to change.

angle The angle in the complex plane [rad].

Returns

A reference to the changed variable.

3.10.2.6 double angle (

const mha_complex_t &self )[inline]

Computes the angle of a complex number in the complex plane.

3.10 Complex arithmetics in the openMHA 65

Parameters

self The complex number whose angle is needed.

Returns

The angle of a complex number in the complex plane.

3.10.2.7 mha_complex_t& operator+= (

mha_complex_t &self,

const mha_complex_t &other )[inline]

Addition of two complex numbers, overwriting the ﬁrst.

3.10.2.8 mha_complex_t operator+ (

const mha_complex_t &self,

const mha_complex_t &other )[inline]

Addition of two complex numbers, result is a temporary object.

3.10.2.9 mha_complex_t& operator+= (

mha_complex_t &self,

mha_real_t other_real )[inline]

Addition of a complex and a real number, overwriting the complex.

3.10.2.10 mha_complex_t operator+ (

const mha_complex_t &self,

mha_real_t other_real )[inline]

Addition of a complex and a real number, result is a temporary object.

3.10.2.11 mha_complex_t& operator-= (

mha_complex_t &self,

const mha_complex_t &other )[inline]

Subtraction of two complex numbers, overwriting the ﬁrst.

3.10.2.12 mha_complex_t operator- (

const mha_complex_t &self,

const mha_complex_t &other )[inline]

Subtraction of two complex numbers, result is a temporary object.

66 CONTENTS

3.10.2.13 mha_complex_t& operator-= (

mha_complex_t &self,

mha_real_t other_real )[inline]

Subtraction of a complex and a real number, overwriting the complex.

3.10.2.14 mha_complex_t operator- (

const mha_complex_t &self,

mha_real_t other_real )[inline]

Subtraction of a complex and a real number, result is a temporary object.

3.10.2.15 mha_complex_t& operator∗= (

mha_complex_t &self,

const mha_complex_t &other )[inline]

Multiplication of two complex numbers, overwriting the ﬁrst.

3.10.2.16 mha_complex_t operator∗(

const mha_complex_t &self,

const mha_complex_t &other )[inline]

Multiplication of two complex numbers, result is a temporary object.

3.10.2.17 mha_complex_t& operator∗= (

mha_complex_t &self,

mha_real_t other_real )[inline]

Multiplication of a complex and a real number, overwriting the complex.

3.10.2.18 mha_complex_t& expi (

mha_complex_t &self,

mha_real_t angle,

mha_real_t factor )[inline]

replaces (!) the value of the given mha_complex_t (p. 431) with a ∗exp(i∗b)

Parameters

self The mha_complex_t (p. 431) variable whose value is about to change.

angle The imaginary exponent.

factor The absolute value of the result.

3.10 Complex arithmetics in the openMHA 67

Returns

A reference to the changed variable.

3.10.2.19 mha_complex_t operator∗(

const mha_complex_t &self,

mha_real_t other_real )[inline]

Multiplication of a complex and a real number, result is a temporary object.

3.10.2.20 mha_real_t abs2 (

const mha_complex_t &self )[inline]

Compute the square of the absolute value of a complex value.

Returns

The square of the absolute value of self.

3.10.2.21 mha_real_t abs (

const mha_complex_t &self )[inline]

Compute the absolute value of a complex value.

Returns

The absolute value of self.

3.10.2.22 mha_complex_t& operator/= (

mha_complex_t &self,

mha_real_t other_real )[inline]

Division of a complex and a real number, overwriting the complex.

3.10.2.23 mha_complex_t operator/ (

const mha_complex_t &self,

mha_real_t other_real )[inline]

Division of a complex and a real number, result is a temporary object.

3.10.2.24 mha_complex_t& safe_div (

mha_complex_t &self,

const mha_complex_t &other,

mha_real_t eps,

mha_real_t eps2 )[inline]

Safe division of two complex numbers, overwriting the ﬁrst.

If abs(divisor) <eps, then divisor is replaced by eps. eps2 = eps∗eps.

68 CONTENTS

3.10.2.25 mha_complex_t& operator/= (

mha_complex_t &self,

const mha_complex_t &other )[inline]

Division of two complex numbers, overwriting the ﬁrst.

3.10.2.26 mha_complex_t operator/ (

const mha_complex_t &self,

const mha_complex_t &other )[inline]

Division of two complex numbers, result is a temporary object.

3.10.2.27 mha_complex_t operator- (

const mha_complex_t &self )[inline]

Unary minus on a complex results in a negative temporary object.

3.10.2.28 bool operator== (

const mha_complex_t &x,

const mha_complex_t &y)[inline]

Compare two complex numbers for equality.

3.10.2.29 bool operator!= (

const mha_complex_t &x,

const mha_complex_t &y)[inline]

Compare two complex numbers for inequality.

3.10.2.30 void conjugate (

mha_complex_t &self )[inline]

Replace (!) the value of this mha_complex_t (p. 431) with its conjugate.

3.10.2.31 mha_complex_t _conjugate (

const mha_complex_t &self )[inline]

Compute the cojugate of this complex value.

Returns

A temporary object holding the conjugate value.

3.10 Complex arithmetics in the openMHA 69

3.10.2.32 void reciprocal (

mha_complex_t &self )[inline]

Replace the value of this complex with its reciprocal.

3.10.2.33 mha_complex_t _reciprocal (

const mha_complex_t &self )[inline]

compute the reciprocal of this complex value.

Returns

A temporary object holding the reciprocal value.

3.10.2.34 void normalize (

mha_complex_t &self )[inline]

Divide a complex by its absolute value, thereby normalizing it (projecting onto the unit circle).

3.10.2.35 void normalize (

mha_complex_t &self,

mha_real_t margin )[inline]

Divide a complex by its absolute value, thereby normalizing it (projecting onto the unit circle),

with a safety margin.

3.10.2.36 bool almost (

const mha_complex_t &self,

const mha_complex_t &other,

mha_real_t times_epsilon = 1e2 )[inline]

Compare two complex numbers for equality except for a small relative error.

Parameters

self The ﬁrst complex number.

other The second complex number.

times_epsilon Permitted relative error is this number multiplied with the machine accuracy

for this Floating point format (std::numeric_limits<mha_real_t>::epsilon)

Returns

true if the relative difference is below times_epsilon ∗std::numeric_limits<mha_real_t>←-

::epsilon

70 CONTENTS

3.10.2.37 bool operator<(

const mha_complex_t &x,

const mha_complex_t &y)[inline]

Compares the absolute values of two complex numbers.

3.11 Fast Fourier Transform functions 71

3.11 Fast Fourier Transform functions

Collaboration diagram for Fast Fourier Transform functions:

The openMHA Toolbox

library Fast Fourier Transform

functions

Typedefs

• typedef void ∗mha_fft_t

Handle for an FFT object.

Functions

•mha_fft_t mha_fft_new (unsigned int n)

Create a new FFT handle.

• void mha_fft_free (mha_fft_t h)

Destroy an FFT handle.

• void mha_fft_wave2spec (mha_fft_t h, const mha_wave_t ∗in, mha_spec_t ∗out)

Tranform waveform segment into spectrum.

• void mha_fft_wave2spec (mha_fft_t h, const mha_wave_t ∗in, mha_spec_t ∗out, bool

swaps)

Tranform waveform segment into spectrum.

• void mha_fft_spec2wave (mha_fft_t h, const mha_spec_t ∗in, mha_wave_t ∗out)

Tranform spectrum into waveform segment.

• void mha_fft_spec2wave (mha_fft_t h, const mha_spec_t ∗in, mha_wave_t ∗out, un-

signed int offset)

Tranform spectrum into waveform segment.

• void mha_fft_forward (mha_fft_t h, mha_spec_t ∗sIn, mha_spec_t ∗sOut)

Complex to complex FFT (forward).

• void mha_fft_backward (mha_fft_t h, mha_spec_t ∗sIn, mha_spec_t ∗sOut)

Complex to complex FFT (backward).

• void mha_fft_forward_scale (mha_fft_t h, mha_spec_t ∗sIn, mha_spec_t ∗sOut)

Complex to complex FFT (forward).

• void mha_fft_backward_scale (mha_fft_t h, mha_spec_t ∗sIn, mha_spec_t ∗sOut)

Complex to complex FFT (backward).

• void mha_fft_wave2spec_scale (mha_fft_t h, const mha_wave_t ∗in, mha_spec_←-

t∗out)

Tranform waveform segment into spectrum.

• void mha_fft_spec2wave_scale (mha_fft_t h, const mha_spec_t ∗in, mha_wave_←-

t∗out)

Tranform spectrum into waveform segment.

72 CONTENTS

3.11.1 Detailed Description

3.11.2 Typedef Documentation

3.11.2.1 typedef void∗mha_fft_t

Handle for an FFT object.

This FFT object is used by the functions mha_fft_wave2spec and mha_fft_spec2wave. The F←-

FT back-end is the FFTW library. The back-end is completely hidden, including external header

ﬁles or linking external libraries is not required.

3.11.3 Function Documentation

3.11.3.1 mha_fft_t mha_fft_new (

unsigned int n)

Create a new FFT handle.

Parameters

nFFT length.

Create a new FFT handle.

Parameters

nFFT length

Return values

FFT object

3.11.3.2 void mha_fft_free (

mha_fft_t h)

Destroy an FFT handle.

Parameters

hHandle to be destroyed.

Destroy an FFT handle.

3.11 Fast Fourier Transform functions 73

Parameters

hFFT object to be removed

3.11.3.3 void mha_fft_wave2spec (

mha_fft_t h,

const mha_wave_t ∗in,

mha_spec_t ∗out )

Tranform waveform segment into spectrum.

Parameters

hFFT handle.

in Input waveform segment.

out Output spectrum.

Tranform waveform segment into spectrum.

Parameters

hFFT object handle

in pointer to input waveform signal

out pointer to output spectrum signal (has to be allocated)

3.11.3.4 void mha_fft_wave2spec (

mha_fft_t h,

const mha_wave_t ∗in,

mha_spec_t ∗out,

bool swaps )

Tranform waveform segment into spectrum.

Like normal wave2spec, but swaps wave buffer halves before transforming if the swaps param-

eter is true.

Warning: These openMHA FFTs adopt a nonstandard scaling scheme in which the forward

transform scales by 1/N and the backward does not scale. We would recommend using the

'_scale' methods instead.

Parameters

hFFT handle.

in Input waveform segment.

out Output spectrum.

swaps Function swaps the ﬁrst and second half of the waveform buffer before the FFT

transform when this parameter is set to true.

74 CONTENTS

3.11.3.5 void mha_fft_spec2wave (

mha_fft_t h,

const mha_spec_t ∗in,

mha_wave_t ∗out )

Tranform spectrum into waveform segment.

Warning: These openMHA FFTs adopt a nonstandard scaling scheme in which the forward

transform scales by 1/N and the backward does not scale. We would recommend using the

'_scale' methods instead.

Parameters

hFFT handle.

in Input spectrum.

out Output waveform segment.

Tranform spectrum into waveform segment.

Parameters

hFFT object handle

in pointer to input spectrum

out pointer to output waveform signal (has to be allocated)

3.11.3.6 void mha_fft_spec2wave (

mha_fft_t h,

const mha_spec_t ∗in,

mha_wave_t ∗out,

unsigned int offset )

Tranform spectrum into waveform segment.

out may have fewer number of frames than needed for a complete iFFT. Only as many frames

are written into out as ﬁt, starting with offset offset of the complete iFFT.

Warning: These openMHA FFTs adopt a nonstandard scaling scheme in which the forward

transform scales by 1/N and the backward does not scale. We would recommend using the

'_scale' methods instead.

Parameters

hFFT handle.

in Input spectrum.

out Output waveform segment.

offset Offset into iFFT wave buffer

3.11 Fast Fourier Transform functions 75

Tranform spectrum into waveform segment.

Only part of the iFFT is tranferred into the out buffer.

Out may have fewer number of freames than needed for a complete iFFT. Only as many frames

are written into out as ﬁt, starting with offset offset of the complete iFFT.

Parameters

hFFT object handle

in pointer to input spectrum

out pointer to output waveform signal (has to be allocated)

offset Offset into complete iFFT buffer.

3.11.3.7 void mha_fft_forward (

mha_fft_t h,

mha_spec_t ∗sIn,

mha_spec_t ∗sOut )

Complex to complex FFT (forward).

sIn and sOut need to have nfft bins (please note that mha_spec_t (p. 465) typically has nfft/2+1

bins for half-complex representation).

Warning: These openMHA FFTs adopt a nonstandard scaling scheme in which the forward

transform scales by 1/N and the backward does not scale. We would recommend using the

'_scale' methods instead.

Parameters

hFFT handle.

sIn Input spectrum.

sOut Output spectrum.

3.11.3.8 void mha_fft_backward (

mha_fft_t h,

mha_spec_t ∗sIn,

mha_spec_t ∗sOut )

Complex to complex FFT (backward).

sIn and sOut need to have nfft bins (please note that mha_spec_t (p. 465) typically has nfft/2+1

bins for half-complex representation).

Warning: These openMHA FFTs adopt a nonstandard scaling scheme in which the forward

transform scales by 1/N and the backward does not scale. We would recommend using the

'_scale' methods instead.

76 CONTENTS

Parameters

hFFT handle.

sIn Input spectrum.

sOut Output spectrum.

3.11.3.9 void mha_fft_forward_scale (

mha_fft_t h,

mha_spec_t ∗sIn,

mha_spec_t ∗sOut )

Complex to complex FFT (forward).

sIn and sOut need to have nfft bins (please note that mha_spec_t (p. 465) typically has nfft/2+1

bins for half-complex representation).

The _scale methods use standard DFT scaling: There is no scaling in the forward transforma-

tion, and 1/N scaling for the backward.

Parameters

hFFT handle.

sIn Input spectrum.

sOut Output spectrum.

3.11.3.10 void mha_fft_backward_scale (

mha_fft_t h,

mha_spec_t ∗sIn,

mha_spec_t ∗sOut )

Complex to complex FFT (backward).

sIn and sOut need to have nfft bins (please note that mha_spec_t (p. 465) typically has nfft/2+1

bins for half-complex representation).

The _scale methods use standard DFT scaling: There is no scaling in the forward transforma-

tion, and 1/N scaling for the backward.

Parameters

hFFT handle.

sIn Input spectrum.

sOut Output spectrum.

3.11 Fast Fourier Transform functions 77

3.11.3.11 void mha_fft_wave2spec_scale (

mha_fft_t h,

const mha_wave_t ∗in,

mha_spec_t ∗out )

Tranform waveform segment into spectrum.

The _scale methods use standard DFT scaling: There is no scaling in the forward transforma-

tion, and 1/N scaling for the backward.

Parameters

hFFT handle.

in Input waveform segment.

out Output spectrum.

3.11.3.12 void mha_fft_spec2wave_scale (

mha_fft_t h,

const mha_spec_t ∗in,

mha_wave_t ∗out )

Tranform spectrum into waveform segment.

The _scale methods use standard DFT scaling: There is no scaling in the forward transforma-

tion, and 1/N scaling for the backward.

Parameters

hFFT handle.

in Input spectrum.

out Output waveform segment.

78 CONTENTS

4 Namespace Documentation

4.1 ac2lsl Namespace Reference

All types for the ac2lsl (p. 78) plugins live in this namespace.

Classes

• class ac2lsl_t

Plugin class of ac2lsl (p. 78).

• class cfg_t

Runtime conﬁguration class of the ac2lsl (p. 78) plugin.

• class save_var_base_t

Interface for ac to lsl bridge variable.

• class save_var_t

Implementation for all ac to lsl bridges except complex types.

• class save_var_t<mha_complex_t >

Template specialization of the ac2lsl (p. 78) bridge to take care of complex numbers.

• struct type_info

Variables

• const std::map<int, type_info >types

4.1.1 Detailed Description

All types for the ac2lsl (p. 78) plugins live in this namespace.

4.1.2 Variable Documentation

4.1.2.1 const std::map<int, type_info>ac2lsl::types

4.2 acmon Namespace Reference

Namespace for displaying ac variables as parser monitors.

Classes

• class ac_monitor_t

A class for converting AC variables to Parser monitors of correct type.

• class acmon_t

4.3 acsave Namespace Reference 79

4.2.1 Detailed Description

Namespace for displaying ac variables as parser monitors.

4.3 acsave Namespace Reference

Classes

• class acsave_t

• class cfg_t

• struct mat4head_t

• class save_var_t

4.4 addsndﬁle Namespace Reference

Classes

• class addsndﬁle_if_t

• class level_adapt_t

• class resampled_soundﬁle_t

Reads sound from ﬁle and resamples it if necessary and wanted.

• class sndﬁle_t

• class waveform_proxy_t

Class helps to specify which instance of MHASignal_waveform_t parent instance is meant in

resampled_soundﬁle_t (p. 184).

Typedefs

• typedef MHAPlugin::conﬁg_t<level_adapt_t >level_adaptor

• typedef MHAPlugin::plugin_t<sndﬁle_t >wave_reader

Enumerations

Functions

• static unsigned resampled_num_frames (unsigned num_source_frames, ﬂoat source←-

_rate, ﬂoat target_rate, addsndﬁle_resampling_mode_t resampling_mode)

80 CONTENTS

4.4.1 Typedef Documentation

4.4.1.1 typedef MHAPlugin::conﬁg_t<level_adapt_t>addsndﬁle::level_adaptor

4.4.1.2 typedef MHAPlugin::plugin_t<sndﬁle_t>addsndﬁle::wave_reader

4.4.2 Enumeration Type Documentation

4.4.2.1 enum addsndﬁle::addsndﬁle_resampling_mode_t

Speciﬁes the resampling mode in resampled_soundﬁle_t (p. 184).

Enumerator

DONT_RESAMPLE_PERMISSIVE

DONT_RESAMPLE_STRICT Do not resample, if the sample rate of the MHA differs

from the sample rate of the sound ﬁle, raise an error.

DO_RESAMPLE Resample.

4.4.3 Function Documentation

4.4.3.1 static unsigned addsndﬁle::resampled_num_frames (

unsigned num_source_frames,

ﬂoat source_rate,

ﬂoat target_rate,

addsndﬁle_resampling_mode_t resampling_mode )[static]

4.5 ADM Namespace Reference

Classes

• class ADM

Adaptive differential microphone, working for speech frequency range.

• class Delay

A delay-line class which can also do subsample-delays for a limited frequency range below

fs/4.

• class Linearphase_FIR

An efﬁcient linear-phase ﬁr ﬁlter implementation.

Functions

• static double subsampledelay_coeff (double samples, double f_design, double fs=1.0)

compute IIR coefﬁcient for subsample delay

4.6 AuditoryProﬁle Namespace Reference 81

Variables

• const double PI = 3.14159265358979312

• const double C= 340

• const double DELAY_FREQ = 2000

• const double START_BETA = 0.5

4.5.1 Function Documentation

4.5.1.1 static double ADM::subsampledelay_coeff (

double samples,

double f_design,

double fs = 1.0 )[static]

compute IIR coefﬁcient for subsample delay

Parameters

samples Constraint: 0.0 <= samples <1.0; Amount of sub-sample delay

f_design design frequency (subsample delay is accurate for this frequency)

fs sampling rate

Returns

IIR coefﬁcient for subsample delay

4.5.2 Variable Documentation

4.5.2.1 const double ADM::PI = 3.14159265358979312

4.5.2.2 const double ADM::C = 340

4.5.2.3 const double ADM::DELAY_FREQ = 2000

4.5.2.4 const double ADM::START_BETA = 0.5

4.6 AuditoryProﬁle Namespace Reference

Namespace for classes and functions around the auditory proﬁle (e.g., audiogram handling)

82 CONTENTS

Classes

• class fmap_t

A class to store frequency dependent data (e.g., HTL and UCL).

• class parser_t

Class to make the auditory proﬁle accessible through the parser interface.

• class proﬁle_t

The Auditory Proﬁle class.

4.6.1 Detailed Description

Namespace for classes and functions around the auditory proﬁle (e.g., audiogram handling)

The auditory proﬁle as deﬁned by HearCom or BMBF Modellbasierte Hoergeraete is stored in

the class AuditoryProﬁle::proﬁle_t (p. 226). Until a complete deﬁnition is available, only the

currently needed elements are implemented.

4.7 coherence Namespace Reference

Classes

• class cohﬂt_if_t

• class cohﬂt_t

• class vars_t

Functions

• void getcipd (mha_complex_t &c, mha_real_t &a, const mha_complex_t &xl, const

mha_complex_t &xr)

4.7.1 Function Documentation

4.7.1.1 void coherence::getcipd (

mha_complex_t &c,

mha_real_t &a,

const mha_complex_t &xl,

const mha_complex_t &xr )[inline]

4.8 dc Namespace Reference

Classes

• class dc_if_t

• class dc_t

• class dc_vars_t

• class dc_vars_validator_t

• class wb_inhib_cfg_t

• class wideband_inhib_vars_t

4.9 dc_simple Namespace Reference 83

Functions

• unsigned int get_audiochannels (unsigned int totalchannels, std::string acname, algo←-

_comm_t ac)

4.8.1 Function Documentation

4.8.1.1 unsigned int dc::get_audiochannels (

unsigned int totalchannels,

std::string acname,

algo_comm_t ac )

4.9 dc_simple Namespace Reference

Classes

• class dc_if_t

• class dc_t

• class dc_vars_t

• class dc_vars_validator_t

• class level_smoother_t

Typedefs

• typedef MHAPlugin::plugin_t<dc_t >DC

• typedef MHAPlugin::conﬁg_t<level_smoother_t >LEVEL

Functions

• void test_fail (const std::vector<ﬂoat >&v, unsigned int s, const std::string &name)

• std::vector<ﬂoat >force_resize (const std::vector<ﬂoat >&v, unsigned int s, const

std::string &name)

•mha_real_t not_zero (mha_real_t x, const std::string &comment="")

84 CONTENTS

4.9.1 Typedef Documentation

4.9.1.1 typedef MHAPlugin::plugin_t<dc_t>dc_simple::DC

4.9.1.2 typedef MHAPlugin::conﬁg_t<level_smoother_t>dc_simple::LEVEL

4.9.2 Function Documentation

4.9.2.1 void dc_simple::test_fail (

const std::vector<ﬂoat >&v,

unsigned int s,

const std::string & name )

4.9.2.2 std::vector<ﬂoat>dc_simple::force_resize (

const std::vector<ﬂoat >&v,

unsigned int s,

const std::string & name )

4.9.2.3 mha_real_t dc_simple::not_zero (

mha_real_t x,

const std::string & comment = "" )

4.10 delay Namespace Reference

Classes

• class interface_t

4.11 delaysum Namespace Reference

This namespace contains the delaysum plugin.

Classes

• class delaysum_if_t

Interface class for the delaysum plugin.

• class delaysum_t

Runtime conﬁguration of the delaysum plugin.

4.11.1 Detailed Description

This namespace contains the delaysum plugin.

4.12 DynComp Namespace Reference 85

4.12 DynComp Namespace Reference

dynamic compression related classes and functions

Classes

• class dc_afterburn_rt_t

Real-time class for after burn effect.

• class dc_afterburn_t

Afterburn class, to be deﬁned as a member of compressors.

• class dc_afterburn_vars_t

Variables for dc_afterburn_t (p. 294) class.

• class gaintable_t

Gain table class.

Functions

•mha_real_t interp1 (const std::vector<mha_real_t >&vX, const std::vector<mha_←-

real_t >&vY, mha_real_t X)

One-dimensional linear interpolation.

•mha_real_t interp2 (const std::vector<mha_real_t >&vX, const std::vector<mha←-

_real_t >&vY, const std::vector<std::vector<mha_real_t > > &mZ, mha_real_t X,

mha_real_t Y)

Linear interpolation in a two-dimensional ﬁeld.

4.12.1 Detailed Description

dynamic compression related classes and functions

4.12.2 Function Documentation

4.12.2.1 mha_real_t DynComp::interp1 (

const std::vector<mha_real_t >&vX,

const std::vector<mha_real_t >&vY,

mha_real_t X)

One-dimensional linear interpolation.

Parameters

vX Vector with input samples.

vY Vector with values at input samples.

XInput value to be interpolated.

86 CONTENTS

Return values

Interpolated value Y(X) at position X.

4.12.2.2 mha_real_t DynComp::interp2 (

const std::vector<mha_real_t >&vX,

const std::vector<mha_real_t >&vY,

const std::vector<std::vector<mha_real_t > > &mZ,

mha_real_t X,

mha_real_t Y)

Linear interpolation in a two-dimensional ﬁeld.

Parameters

vX Vector with input samples, ﬁrst dimension.

vY Vector with input samples, second dimension.

mZ Field with values at input samples.

XFirst dimension of input value to be interpolated.

YSecond dimension of input value to be interpolated.

Return values

Interpolated value Z(X,Y) at position X,Y.

4.13 fader_wave Namespace Reference

Classes

• class fader_wave_if_t

• class level_adapt_t

Typedefs

• typedef MHAPlugin::plugin_t<level_adapt_t >level_adaptor

4.13.1 Typedef Documentation

4.13.1.1 typedef MHAPlugin::plugin_t<level_adapt_t>fader_wave::level_adaptor

4.14 fftfbpow Namespace Reference

Namespace for the fftfbpow plugin.

4.15 fftﬁlterbank Namespace Reference 87

Classes

• class fftfbpow_interface_t

Interface class for fftfbpow plugin.

• class fftfbpow_t

Run time conﬁguration for the fftfbpow plugin.

4.14.1 Detailed Description

Namespace for the fftfbpow plugin.

4.15 fftﬁlterbank Namespace Reference

Classes

• class fftfb_interface_t

• class fftfb_plug_t

4.16 fshift Namespace Reference

All types for the fshift plugin live in this namespace.

Classes

• class fshift_conﬁg_t

fshift runtime conﬁg class

• class fshift_t

fshift plugin interface class

Functions

• int fft_ﬁnd_bin (mha_real_t frequency, unsigned fftlen, mha_real_t srate)

Finds bin number of FFT bin nearest to the given frequency.

4.16.1 Detailed Description

All types for the fshift plugin live in this namespace.

4.16.2 Function Documentation

4.16.2.1 int fshift::fft_ﬁnd_bin (

mha_real_t frequency,

unsigned fftlen,

mha_real_t srate )

Finds bin number of FFT bin nearest to the given frequency.

88 CONTENTS

Parameters

frequency The frequency for which to look. Has to be in range [-srate/2,+srate/2]

fftlen Length of the FFT.

srate Sampling rate of the waveform from which the FFT originates.

Returns

Bin number of the FFT bin corresponding to frequency

4.17 fshift_hilbert Namespace Reference

All types for the hilbert frequency shifter live in this namespace.

Classes

• class frequency_translator_t

• class hilbert_shifter_t

4.17.1 Detailed Description

All types for the hilbert frequency shifter live in this namespace.

4.18 gain Namespace Reference

Classes

• class gain_if_t

• class scaler_t

4.19 gtfb_analyzer Namespace Reference

Classes

• struct gtfb_analyzer_cfg_t

Conﬁguration for Gammatone Filterbank Analyzer.

• class gtfb_analyzer_t

Gammatone Filterbank Analyzer Plugin.

4.20 matrixmixer Namespace Reference 89

4.20 matrixmixer Namespace Reference

Classes

• class cfg_t

• class matmix_t

4.21 MHA_AC Namespace Reference

Functions and classes for Algorithm Communication (AC) support.

Classes

• class ac2matrix_helper_t

• class ac2matrix_t

Copy AC variable to a matrix.

• class acspace2matrix_t

Copy all or a subset of all numeric AC variables into an array of matrixes.

• class double_t

Insert a double precision ﬂoating point variable into the AC space.

• class ﬂoat_t

Insert a ﬂoat point variable into the AC space.

• class int_t

Insert a integer variable into the AC space.

• class spectrum_t

Insert a MHASignal::spectrum_t (p. 800) class into the AC space.

• class stat_t

• class waveform_t

Insert a MHASignal::waveform_t (p. 812) class into the AC space.

Functions

•mha_spec_t get_var_spectrum (algo_comm_t ac, const std::string &name)

Convert an AC variable into a spectrum.

•mha_wave_t get_var_waveform (algo_comm_t ac, const std::string &name)

Convert an AC variable into a waveform.

• int get_var_int (algo_comm_t ac, const std::string &name)

Return value of an integer scalar AC variable.

• ﬂoat get_var_ﬂoat (algo_comm_t ac, const std::string &name)

Return value of an ﬂoating point scalar AC variable.

• std::vector<ﬂoat >get_var_vﬂoat (algo_comm_t ac, const std::string &name)

Return value of an ﬂoating point vector AC variable as standard vector of ﬂoats.

90 CONTENTS

4.21.1 Detailed Description

Functions and classes for Algorithm Communication (AC) support.

4.22 mha_error_helpers Namespace Reference

Functions

• unsigned digits (unsigned n)

Compute number of decimal digits required to represent an unsigned integer.

• unsigned snprintf_required_length (const char ∗formatstring,...)

snprintf_required_length Compute the number of bytes (excluding the terminating nul) required

to store the result of an snprintf.

4.22.1 Function Documentation

4.22.1.1 unsigned mha_error_helpers::digits (

unsigned n)

Compute number of decimal digits required to represent an unsigned integer.

Parameters

nThe unsigned integer that we want to know the number of required decimal digits for.

return The number of decimal digits in n.

4.22.1.2 unsigned mha_error_helpers::snprintf_required_length (

const char ∗formatstring,

... )

snprintf_required_length Compute the number of bytes (excluding the terminating nul) required

to store the result of an snprintf.

Parameters

formatstring The format string with standard printf formatstring

Returns

the number of bytes required by printf without the terminating nul

4.23 MHA_TCP Namespace Reference 91

4.23 MHA_TCP Namespace Reference

A Namespace for TCP helper classes.

Classes

• class Async_Notify

Portable Multiplexable cross-thread notiﬁcation.

• class Client

A portable class for a tcp client connections.

• class Connection

Connection (p. 471) handles Communication between client and server, is used on both sides.

• class Event_Watcher

OS-independent event watcher, uses select on Unix and WaitForMultipleObjects on Windows.

• struct OS_EVENT_TYPE

• class Server

• class sock_init_t

• class Sockaccept_Event

• class Sockread_Event

Watch socket for incoming data.

• class Sockwrite_Event

• class Thread

A very simple class for portable threads.

• class Timeout_Event

• class Timeout_Watcher

OS-independent event watcher with internal ﬁxed-end-time timeout.

• class Wakeup_Event

A base class for asynchronous wakeup events.

Typedefs

• typedef int SOCKET

Functions

• std::string STRERROR (int err)

Portable conversion from error number to error string.

• std::string HSTRERROR (int err)

Portable conversion from hostname error number to error string.

• int N_ERRNO ()

Portable access to last network error number.

• int H_ERRNO ()

Portable access to last hostname error number.

92 CONTENTS

• int G_ERRNO ()

Portable access to last non-network error number.

• double dtime ()

Time access function for system's high resolution time, retrieve current time as double.

• double dtime (const struct timeval &tv)

Time access function for unix' high resolution time, converts struct timeval to double.

• struct timeval stime (double d)

Time access function for unix' high resolution time, converts time from double to struct timeval.

Variables

• class MHA_TCP::sock_init_t sock_initializer

4.23.1 Detailed Description

A Namespace for TCP helper classes.

4.23.2 Typedef Documentation

4.23.2.1 typedef int MHA_TCP::SOCKET

4.23.3 Function Documentation

4.23.3.1 std::string MHA_TCP::STRERROR (

int err )

Portable conversion from error number to error string.

4.23.3.2 std::string MHA_TCP::HSTRERROR (

int err )

Portable conversion from hostname error number to error string.

4.23.3.3 int MHA_TCP::N_ERRNO ( )

Portable access to last network error number.

4.23.3.4 int MHA_TCP::H_ERRNO ( )

Portable access to last hostname error number.

4.24 mhachain Namespace Reference 93

4.23.3.5 int MHA_TCP::G_ERRNO ( )

Portable access to last non-network error number.

4.23.3.6 double MHA_TCP::dtime ( )

Time access function for system's high resolution time, retrieve current time as double.

4.23.3.7 double MHA_TCP::dtime (

const struct timeval & tv )

Time access function for unix' high resolution time, converts struct timeval to double.

4.23.3.8 struct timeval MHA_TCP::stime (

double d)

Time access function for unix' high resolution time, converts time from double to struct timeval.

4.23.4 Variable Documentation

4.23.4.1 class MHA_TCP::sock_init_t MHA_TCP::sock_initializer

4.24 mhachain Namespace Reference

Classes

• class chain_base_t

• class mhachain_t

• class plugs_t

4.25 MHAEvents Namespace Reference

Collection of event handling classes.

Classes

• class connector_base_t

• class connector_t

• class emitter_t

Class for emitting openMHA events.

• class patchbay_t

Patchbay which connects any event emitter with any member function of the parameter class.

94 CONTENTS

4.25.1 Detailed Description

Collection of event handling classes.

4.26 MHAFilter Namespace Reference

Namespace for IIR and FIR ﬁlter classes.

Classes

• class adapt_ﬁlter_param_t

• class adapt_ﬁlter_state_t

• class adapt_ﬁlter_t

Adaptive ﬁlter.

• class blockprocessing_polyphase_resampling_t

A class that does polyphase resampling and takes into account block processing.

• class complex_bandpass_t

Complex bandpass ﬁlter.

• class diff_t

Differentiator class (non-normalized)

• class fftﬁlter_t

FFT based FIR ﬁlter implementation.

• class fftﬁlterbank_t

FFT based FIR ﬁlterbank implementation.

• class ﬁlter_t

Generic IIR ﬁlter class.

• class gamma_ﬂt_t

Class for gammatone ﬁlter.

• class iir_ﬁlter_state_t

• class iir_ﬁlter_t

IIR ﬁlter class wrapper for integration into parser structure.

• class iir_ord1_real_t

First order recursive ﬁlter.

• class o1_ar_ﬁlter_t

First order attack-release lowpass ﬁlter.

• class o1ﬂt_lowpass_t

First order low pass ﬁlter.

• class o1ﬂt_maxtrack_t

First order maximum tracker.

• class o1ﬂt_mintrack_t

First order minimum tracker.

• class partitioned_convolution_t

A ﬁlter class for partitioned convolution.

4.26 MHAFilter Namespace Reference 95

• class polyphase_resampling_t

A class that performs polyphase resampling.

• class resampling_ﬁlter_t

Hann shaped low pass ﬁlter for resampling.

• class smoothspec_t

Smooth spectral gains, create a windowed impulse response.

• class thirdoctave_analyzer_t

• struct transfer_function_t

a structure containing a source channel number, a target channel number, and an impulse

response.

• struct transfer_matrix_t

A sparse matrix of transfer function partitionss.

Functions

• void make_friendly_number (mha_real_t &x)

• void make_friendly_number (mha_complex_t &x)

• void make_friendly_number (double &x)

• void o1_lp_coeffs (const mha_real_t tau, const mha_real_t fs, mha_real_t &c1, mha←-

_real_t &c2)

Set ﬁrst order ﬁlter coefﬁcients from time constant and sampling rate.

• void butter_stop_ord1 (double ∗A, double ∗B, double f1, double f2, double fs)

Setup a ﬁrst order butterworth band stop ﬁlter.

•MHASignal::waveform_t ∗spec2ﬁr (const mha_spec_t ∗spec, const unsigned int fftlen,

const MHAWindow::base_t &window, const bool minphase)

Create a windowed impulse response/FIR ﬁlter coefﬁcients from a spectrum.

• unsigned gcd (unsigned a, unsigned b)

greatest common divisor

• double sinc (double x)

sin(x)/x function, coping with x=0.

• std::pair<unsigned, unsigned >resampling_factors (ﬂoat source_sampling_rate, ﬂoat

target_sampling_rate, ﬂoat factor=1.0f)

Computes rational resampling factor from two sampling rates.

4.26.1 Detailed Description

Namespace for IIR and FIR ﬁlter classes.

96 CONTENTS

4.26.2 Function Documentation

4.26.2.1 void MHAFilter::make_friendly_number (

mha_real_t &x)[inline]

4.26.2.2 void MHAFilter::make_friendly_number (

mha_complex_t &x)[inline]

4.26.2.3 void MHAFilter::make_friendly_number (

double & x)[inline]

4.26.2.4 void MHAFilter::o1_lp_coeffs (

const mha_real_t tau,

const mha_real_t fs,

mha_real_t &c1,

mha_real_t &c2 )

Set ﬁrst order ﬁlter coefﬁcients from time constant and sampling rate.

Parameters

tau Time constant

fs Sampling rate

Return values

c1 Recursive ﬁlter coefﬁcient

c2 Non-recursive ﬁlter coefﬁcient

4.26.2.5 void MHAFilter::butter_stop_ord1 (

double ∗A,

double ∗B,

double f1,

double f2,

double fs )

Setup a ﬁrst order butterworth band stop ﬁlter.

This function calculates the ﬁlter coefﬁcients of a ﬁrst order butterworth band stop ﬁlter.

Return values

Arecursive ﬁlter coefﬁcients

Bnon recursive ﬁlter coefﬁcients

4.26 MHAFilter Namespace Reference 97

Parameters

f1 lower frequency

f2 upper frequency

fs sample frequency

4.26.2.6 MHASignal::waveform_t ∗MHAFilter::spec2ﬁr (

const mha_spec_t ∗spec,

const unsigned int fftlen,

const MHAWindow::base_t &window,

const bool minphase )

Create a windowed impulse response/FIR ﬁlter coefﬁcients from a spectrum.

Parameters

spec Input spectrum

fftlen FFT length of spectrum

window Window shape (with length, e.g. initialized with MHAWindow::hanning(54)).

minphase Flag, true if original phase should be discarded and replaced by a minimal

phase function.

4.26.2.7 unsigned MHAFilter::gcd (

unsigned a,

unsigned b)[inline]

greatest common divisor

4.26.2.8 double MHAFilter::sinc (

double x)

sin(x)/x function, coping with x=0.

This is the historical sinc function, not the normalized sinc function.

4.26.2.9 std::pair<unsigned, unsigned >MHAFilter::resampling_factors (

ﬂoat source_sampling_rate,

ﬂoat target_sampling_rate,

ﬂoat factor = 1.0f )

Computes rational resampling factor from two sampling rates.

The function will fail if either sampling_rate ∗factor is not an integer

98 CONTENTS

Parameters

source_sampling_rate The original sampling rate

target_sampling_rate The desired sampling rate

factor A helper factor to use for non-integer sampling rates

Returns

a pair that contains ﬁrst the upsampling factor and second the downsampling factor re-

quired for the speciﬁed resampling.

Exceptions

MHA_Error (p. 445)if no rational resampling factor can be found.

4.27 MHAIOJack Namespace Reference

JACK IO.

Classes

• class io_jack_t

Main class for JACK IO.

4.27.1 Detailed Description

JACK IO.

4.28 MHAIOJackdb Namespace Reference

Classes

• class io_jack_t

Main class for JACK IO.

4.29 MHAIOPortAudio Namespace Reference

Classes

• class device_info_t

• class io_portaudio_t

Main class for Portaudio sound IO.

4.30 MHAJack Namespace Reference 99

Functions

• static std::string parserFriendlyName (const std::string &in)

4.29.1 Function Documentation

4.29.1.1 static std::string MHAIOPortAudio::parserFriendlyName (

const std::string & in )[static]

4.30 MHAJack Namespace Reference

Classes and functions for openMHA and JACK interaction.

Classes

• class client_avg_t

Generic JACK client for averaging a system response across time.

• class client_noncont_t

Generic client for synchronous playback and recording of waveform fragments.

• class client_t

Generic asynchronous JACK client.

• class port_t

Class for one channel/port.

Functions

• void io (mha_wave_t ∗s_out, mha_wave_t ∗s_in, const std::string &name, const std←-

::vector<std::string >&p_out, const std::vector<std::string >&p_in, ﬂoat ∗srate=NULL,

unsigned int ∗fragsize=NULL, bool use_jack_transport=false)

Functional form of generic client for synchronous playback and recording of waveform frag-

ments.

• std::vector<unsigned int >get_port_capture_latency (const std::vector<std::string >

&ports)

Return the JACK port latency of ports.

• std::vector<int >get_port_capture_latency_int (const std::vector<std::string >

&ports)

Return the JACK port latency of ports.

• std::vector<unsigned int >get_port_playback_latency (const std::vector<std::string

>&ports)

Return the JACK port latency of ports.

• std::vector<int >get_port_playback_latency_int (const std::vector<std::string >

&ports)

100 CONTENTS

4.30.1 Detailed Description

Classes and functions for openMHA and JACK interaction.

4.30.2 Function Documentation

4.30.2.1 void MHAJack::io (

mha_wave_t ∗s_out,

mha_wave_t ∗s_in,

const std::string & name,

const std::vector<std::string >&p_out,

const std::vector<std::string >&p_in,

ﬂoat ∗srate = NULL,

unsigned int ∗fragsize = NULL,

bool use_jack_transport = false )

Functional form of generic client for synchronous playback and recording of waveform frag-

ments.

4.30.2.2 std::vector<unsigned int >MHAJack::get_port_capture_latency (

const std::vector<std::string >&ports )

Return the JACK port latency of ports.

Parameters

ports Ports to be tested

Returns

Latency vector (one entry for each port)

4.30.2.3 std::vector<int >MHAJack::get_port_capture_latency_int (

const std::vector<std::string >&ports )

Return the JACK port latency of ports.

Parameters

ports Ports to be tested

4.31 MHAKernel Namespace Reference 101

Returns

Latency vector (one entry for each port)

4.30.2.4 std::vector<unsigned int >MHAJack::get_port_playback_latency (

const std::vector<std::string >&ports )

Return the JACK port latency of ports.

Parameters

ports Ports to be tested

Returns

Latency vector (one entry for each port)

4.30.2.5 std::vector<int >MHAJack::get_port_playback_latency_int (

const std::vector<std::string >&ports )

4.31 MHAKernel Namespace Reference

Classes

• class algo_comm_class_t

• class comm_var_map_t

Functions

•algo_comm_class_t ∗algo_comm_safe_cast (void ∗)

4.31.1 Function Documentation

4.31.1.1 MHAKernel::algo_comm_class_t ∗MHAKernel::algo_comm_safe_cast (

void ∗h)

4.32 MHAMultiSrc Namespace Reference

Collection of classes for selecting audio chunks from multiple sources.

102 CONTENTS

Classes

• class base_t

Base class for source selection.

• class channel_t

• class channels_t

• class spectrum_t

• class waveform_t

4.32.1 Detailed Description

Collection of classes for selecting audio chunks from multiple sources.

4.33 MHAOvlFilter Namespace Reference

Namespace for overlapping FFT based ﬁlter bank classes and functions.

Namespaces

•barkscale

•FreqScaleFun

Transform functions from linear scale in Hz to new frequency scales.

•ShapeFun

Shape functions for overlapping ﬁlters.

Classes

• class band_descriptor_t

• class fftfb_ac_info_t

• class fftfb_t

FFT based overlapping ﬁlter bank.

• class fftfb_vars_t

Set of conﬁguration variables for FFT-based overlapping ﬁlters.

• class fscale_bw_t

• class fscale_t

• class fspacing_t

Class for frequency spacing, used by ﬁlterbank shape generator class.

• class overlap_save_ﬁlterbank_analytic_t

• class overlap_save_ﬁlterbank_t

A time-domain minimal phase ﬁlter bank with frequency shapes from MHAOvlFilter::fftfb_t

(p. 616).

• class scale_var_t

4.34 MHAOvlFilter::barkscale Namespace Reference 103

Typedefs

• typedef mha_real_t(scale_fun_t) (mha_real_t)

4.33.1 Detailed Description

Namespace for overlapping FFT based ﬁlter bank classes and functions.

4.33.2 Typedef Documentation

4.33.2.1 typedef mha_real_t( MHAOvlFilter::scale_fun_t) (mha_real_t)

4.34 MHAOvlFilter::barkscale Namespace Reference

Classes

• class bark2hz_t

• class hz2bark_t

Variables

•mha_real_t vfreq [BARKSCALE_ENTRIES]

•mha_real_t vbark [BARKSCALE_ENTRIES]

4.34.1 Variable Documentation

4.34.1.1 mha_real_t MHAOvlFilter::barkscale::vfreq

4.34.1.2 mha_real_t MHAOvlFilter::barkscale::vbark

4.35 MHAOvlFilter::FreqScaleFun Namespace Reference

Transform functions from linear scale in Hz to new frequency scales.

104 CONTENTS

Functions

•mha_real_t hz2hz (mha_real_t x)

Dummy scale transformation Hz to Hz.

•mha_real_t hz2khz (mha_real_t x)

•mha_real_t hz2octave (mha_real_t x)

•mha_real_t hz2third_octave (mha_real_t x)

•mha_real_t hz2bark (mha_real_t x)

Transformation to bark scale.

•mha_real_t hz2bark_analytic (mha_real_t)

•mha_real_t hz2erb (mha_real_t)

•mha_real_t hz2erb_glasberg1990 (mha_real_t)

•mha_real_t hz2log (mha_real_t x)

Third octave frequency scale.

•mha_real_t inv_scale (mha_real_t,mha_real_t(∗)(mha_real_t))

4.35.1 Detailed Description

Transform functions from linear scale in Hz to new frequency scales.

4.35.2 Function Documentation

4.35.2.1 mha_real_t MHAOvlFilter::FreqScaleFun::hz2hz (

mha_real_t x)

Dummy scale transformation Hz to Hz.

This function implements a dummy scale transformation (linear frequency scale).

Parameters

xInput frequency in Hz

Returns

Frequency in Hz

4.35.2.2 mha_real_t MHAOvlFilter::FreqScaleFun::hz2khz (

mha_real_t x)

4.35.2.3 mha_real_t MHAOvlFilter::FreqScaleFun::hz2octave (

mha_real_t x)

4.35 MHAOvlFilter::FreqScaleFun Namespace Reference 105

4.35.2.4 mha_real_t MHAOvlFilter::FreqScaleFun::hz2third_octave (

mha_real_t x)

4.35.2.5 mha_real_t MHAOvlFilter::FreqScaleFun::hz2bark (

mha_real_t x)

Transformation to bark scale.

This function implements a critical band rate (bark) scale.

Parameters

xInput frequency in Hz

Returns

Critical band rate in Bark

4.35.2.6 mha_real_t MHAOvlFilter::FreqScaleFun::hz2bark_analytic (

mha_real_t x)

4.35.2.7 mha_real_t MHAOvlFilter::FreqScaleFun::hz2erb (

mha_real_t x)

4.35.2.8 mha_real_t MHAOvlFilter::FreqScaleFun::hz2erb_glasberg1990 (

mha_real_t x)

4.35.2.9 mha_real_t MHAOvlFilter::FreqScaleFun::hz2log (

mha_real_t x)

Third octave frequency scale.

This function implements a third octave scale. Frequencies below 16 Hz are mapped to 16 Hz.

Parameters

xFrequency in Hz

Returns

Third octaves relative to 1000 Hz

4.35.2.10 mha_real_t MHAOvlFilter::FreqScaleFun::inv_scale (

mha_real_t y,

mha_real_t(∗)(mha_real_t)fun )

106 CONTENTS

4.36 MHAOvlFilter::ShapeFun Namespace Reference

Shape functions for overlapping ﬁlters.

Functions

•mha_real_t rect (mha_real_t x)

Filter shape function for rectangular ﬁlters.

•mha_real_t linear (mha_real_t x)

Filter shape function for sawtooth ﬁlters.

•mha_real_t hann (mha_real_t x)

Filter shape function for hanning shaped ﬁlters.

•mha_real_t expﬂt (mha_real_t)

•mha_real_t gauss (mha_real_t)

4.36.1 Detailed Description

Shape functions for overlapping ﬁlters.

4.36.2 Function Documentation

4.36.2.1 mha_real_t MHAOvlFilter::ShapeFun::rect (

mha_real_t x)

Filter shape function for rectangular ﬁlters.

This function creates rectangular ﬁlter shapes. The edge is exactly half way between two center

frequencies (on a given scale).

Parameters

xInput value in the range [-1,1].

Returns

Weigth function in the range [0,1]

4.36.2.2 mha_real_t MHAOvlFilter::ShapeFun::linear (

mha_real_t x)

Filter shape function for sawtooth ﬁlters.

4.37 MHAParser Namespace Reference 107

This function creates sawtooth ﬁlter shapes. They rise linearly form 0 to 1 in the interval from

the lower neighbor center frequency to the band center frequency and from 1 to 0 in the interval

from the band center frequency to the upper neighbour band center frequency. Linear means

linear on a given frequency scale.

Parameters

xInput value in the range [-1,1].

Returns

Weigth function in the range [0,1]

4.36.2.3 mha_real_t MHAOvlFilter::ShapeFun::hann (

mha_real_t x)

Filter shape function for hanning shaped ﬁlters.

This function creates hanning window shaped ﬁlters.

Parameters

xInput value in the range [-1,1].

Returns

Weigth function in the range [0,1]

4.36.2.4 mha_real_t MHAOvlFilter::ShapeFun::expﬂt (

mha_real_t x)

4.36.2.5 mha_real_t MHAOvlFilter::ShapeFun::gauss (

mha_real_t x)

4.37 MHAParser Namespace Reference

Name space for the openMHA-Parser conﬁguration language.

Namespaces

•StrCnv

String converter namespace.

108 CONTENTS

Classes

• class base_t

Base class for all parser items.

• class bool_mon_t

Monitor with string value.

• class bool_t

Variable with a boolean value ("yes"/"no")

• class c_ifc_parser_t

• class commit_t

Parser variable with event-emission functionality.

• class complex_mon_t

Monitor with complex value.

• class complex_t

Variable with complex value.

• class entry_t

• class expression_t

• class ﬂoat_mon_t

Monitor with ﬂoat value.

• class ﬂoat_t

Variable with ﬂoat value.

• class int_mon_t

Monitor variable with int value.

• class int_t

Variable with integer value.

• class keyword_list_t

Keyword list class.

• class kw_t

Variable with keyword list value.

• class mcomplex_mon_t

Matrix of complex numbers monitor.

• class mcomplex_t

Matrix variable with complex value.

• class mﬂoat_mon_t

Matrix of ﬂoats monitor.

• class mﬂoat_t

Matrix variable with ﬂoat value.

• class mhaconﬁg_mon_t

• class mhapluginloader_t

Class to create a plugin loader in a parser, including the load logic.

• class mint_mon_t

Matrix of ints monitor.

• class monitor_t

Base class for monitors and variable nodes.

• class parser_t

4.37 MHAParser Namespace Reference 109

Parser node class.

• class range_var_t

Base class for all variables with a numeric value range.

• class string_mon_t

Monitor with string value.

• class string_t

Variable with a string value.

• class variable_t

Base class for variable nodes.

• class vcomplex_mon_t

Monitor with vector of complex values.

• class vcomplex_t

Vector variable with complex value.

• class vﬂoat_mon_t

Vector of ﬂoats monitor.

• class vﬂoat_t

Vector variable with ﬂoat value.

• class vint_mon_t

Vector of ints monitor.

• class vint_t

Variable with vector<int>value.

• class vstring_mon_t

Vector of monitors with string value.

• class vstring_t

Vector variable with string values.

• class window_t

MHA conﬁguration interface for a window function generator.

Typedefs

• typedef std::string(base_t::∗opact_t) (expression_t &)

• typedef std::string(base_t::∗query_t) (const std::string &)

• typedef std::map<std::string, opact_t >opact_map_t

• typedef std::map<std::string, query_t >query_map_t

• typedef std::list<entry_t >entry_map_t

• typedef int(∗c_parse_cmd_t) (void ∗, const char ∗, char ∗, unsigned int)

• typedef const char ∗(∗c_parse_err_t) (void ∗, int)

110 CONTENTS

Functions

• int get_precision ()

• std::string commentate (const std::string &s)

• void trim (std::string &s)

• std::string cfg_dump (base_t ∗, const std::string &)

• std::string cfg_dump_short (base_t ∗, const std::string &)

• std::string all_dump (base_t ∗, const std::string &)

• std::string mon_dump (base_t ∗, const std::string &)

• std::string all_ids (base_t ∗, const std::string &, const std::string &="")

• void strreplace (std::string &, const std::string &, const std::string &)

string replace function

• void envreplace (std::string &s)

4.37.1 Detailed Description

Name space for the openMHA-Parser conﬁguration language.

This namespace contains all classes which are needed for the implementation of the open←-

MHA conﬁguration language. For details on the script language itself please see section The

openMHA conﬁguration language (p. 34).

4.37.2 List of valid MHAParser items

•Sub-parser:parser_t (p. 688)

•Variables:

Numeric variables: int_t (p. 664), vint_t (p. 713), ﬂoat_t (p. 659), vﬂoat_t (p. 709),

mﬂoat_t (p. 677)

Other variables: string_t (p. 697), vstring_t (p. 717), kw_t (p. 669), bool_t (p. 645)

•Monitors:

Numeric monitors: int_mon_t (p. 662), vint_mon_t (p. 711), ﬂoat_mon_t (p. 658),

vﬂoat_mon_t (p. 707)

mﬂoat_mon_t (p. 676)

mcomplex_mon_t (p. 672)

Other monitors: bool_mon_t (p. 644), string_mon_t (p. 696), vstring_mon_t (p. 715)

Members can be inserted into the conﬁguration namespace by using MHAParser::insert_item()

or the insert_member() (p. 1021) macro.

4.37 MHAParser Namespace Reference 111

4.37.3 Typedef Documentation

4.37.3.1 typedef std::string(base_t::∗MHAParser::opact_t) (expression_t &)

4.37.3.2 typedef std::string(base_t::∗MHAParser::query_t) (const std::string &)

4.37.3.3 typedef std::map<std::string,opact_t>MHAParser::opact_map_t

4.37.3.4 typedef std::map<std::string,query_t>MHAParser::query_map_t

4.37.3.5 typedef std::list<entry_t>MHAParser::entry_map_t

4.37.3.6 typedef int(∗MHAParser::c_parse_cmd_t) (void ∗, const char ∗, char ∗, unsigned int)

4.37.3.7 typedef const char∗(∗MHAParser::c_parse_err_t) (void ∗, int)

4.37.4 Function Documentation

4.37.4.1 int MHAParser::get_precision ( )

4.37.4.2 std::string MHAParser::commentate (

const std::string & s)

4.37.4.3 void MHAParser::trim (

std::string & s)

4.37.4.4 std::string MHAParser::cfg_dump (

base_t ∗p,

const std::string & pref )

4.37.4.5 std::string MHAParser::cfg_dump_short (

base_t ∗p,

const std::string & pref )

4.37.4.6 std::string MHAParser::all_dump (

base_t ∗p,

const std::string & pref )

4.37.4.7 std::string MHAParser::mon_dump (

base_t ∗p,

const std::string & pref )

4.37.4.8 std::string MHAParser::all_ids (

base_t ∗p,

const std::string & pref,

const std::string & id = "" )

4.37.4.9 void MHAParser::strreplace (

std::string & s,

const std::string & arg,

const std::string & rep )

string replace function

112 CONTENTS

Parameters

starget string

arg search pattern

rep replace pattern

4.37.4.10 void MHAParser::envreplace (

std::string & s)

4.38 MHAParser::StrCnv Namespace Reference

String converter namespace.

Functions

• int num_brackets (const std::string &s)

Return number of brackets at beginning and end of string.

• int bracket_balance (const std::string &s)

• void str2val (const std::string &, bool &)

Convert from string.

• void str2val (const std::string &, ﬂoat &)

Convert from string.

• void str2val (const std::string &, mha_complex_t &)

Convert from string.

• void str2val (const std::string &, int &)

Convert from string.

• void str2val (const std::string &, keyword_list_t &)

Convert from string.

• void str2val (const std::string &, std::string &)

Convert from string.

•template<class arg_t >

void str2val (const std::string &s, std::vector<arg_t >&val)

Converter for vector types.

•template<>

void str2val<mha_real_t >(const std::string &s, std::vector<mha_real_t >&v)

Converter for vector<mha_real_t>with Matlab-style expansion.

•template<class arg_t >

void str2val (const std::string &s, std::vector<std::vector<arg_t > > &val)

Converter for matrix types.

• std::string val2str (const bool &)

Convert to string.

• std::string val2str (const ﬂoat &)

Convert to string.

• std::string val2str (const mha_complex_t &)

4.38 MHAParser::StrCnv Namespace Reference 113

Convert to string.

• std::string val2str (const int &)

Convert to string.

• std::string val2str (const keyword_list_t &)

Convert to string.

• std::string val2str (const std::string &)

Convert to string.

• std::string val2str (const std::vector<ﬂoat >&)

Convert to string.

• std::string val2str (const std::vector<mha_complex_t >&)

Convert to string.

• std::string val2str (const std::vector<int >&)

Convert to string.

• std::string val2str (const std::vector<std::vector<int > > &)

Convert to string.

• std::string val2str (const std::vector<std::string >&)

Convert to string.

• std::string val2str (const std::vector<std::vector<ﬂoat > > &)

Convert to string.

• std::string val2str (const std::vector<std::vector<mha_complex_t > > &)

Convert to string.

4.38.1 Detailed Description

String converter namespace.

The functions deﬁned in this namespace manage the conversions from C++ variables to strings

and back. It was tried to keep a matlab compatible string format for vectors and vectors of

vectors.

4.38.2 Function Documentation

4.38.2.1 int MHAParser::StrCnv::num_brackets (

const std::string & s)

Return number of brackets at beginning and end of string.

Parameters

sString

114 CONTENTS

Returns

Number of brackets, or -1 for empty string

4.38.2.2 int MHAParser::StrCnv::bracket_balance (

const std::string & s)

4.38.2.3 void MHAParser::StrCnv::str2val (

const std::string & s,

bool & v)

Convert from string.

4.38.2.4 void MHAParser::StrCnv::str2val (

const std::string & s,

ﬂoat & v)

Convert from string.

4.38.2.5 void MHAParser::StrCnv::str2val (

const std::string & s,

mha_complex_t &v)

Convert from string.

4.38.2.6 void MHAParser::StrCnv::str2val (

const std::string & s,

int & v)

Convert from string.

4.38.2.7 void MHAParser::StrCnv::str2val (

const std::string & s,

MHAParser::keyword_list_t &v)

Convert from string.

4.38.2.8 void MHAParser::StrCnv::str2val (

const std::string & s,

std::string & v)

Convert from string.

4.38.2.9 template<class arg_t >void MHAParser::StrCnv::str2val (

const std::string & s,

std::vector<arg_t >&val )

Converter for vector types.

4.38 MHAParser::StrCnv Namespace Reference 115

4.38.2.10 template<> void MHAParser::StrCnv::str2val<mha_real_t >(

const std::string & s,

std::vector<mha_real_t >&v)

Converter for vector<mha_real_t>with Matlab-style expansion.

4.38.2.11 template<class arg_t >void MHAParser::StrCnv::str2val (

const std::string & s,

std::vector<std::vector<arg_t > > &val )

Converter for matrix types.

4.38.2.12 std::string MHAParser::StrCnv::val2str (

const bool & v)

Convert to string.

4.38.2.13 std::string MHAParser::StrCnv::val2str (

const ﬂoat & v)

Convert to string.

4.38.2.14 std::string MHAParser::StrCnv::val2str (

const mha_complex_t &v)

Convert to string.

4.38.2.15 std::string MHAParser::StrCnv::val2str (

const int & v)

Convert to string.

4.38.2.16 std::string MHAParser::StrCnv::val2str (

const keyword_list_t &v)

Convert to string.

4.38.2.17 std::string MHAParser::StrCnv::val2str (

const std::string & v)

Convert to string.

4.38.2.18 std::string MHAParser::StrCnv::val2str (

const std::vector<ﬂoat >&v)

Convert to string.

116 CONTENTS

4.38.2.19 std::string MHAParser::StrCnv::val2str (

const std::vector<mha_complex_t >&v)

Convert to string.

4.38.2.20 std::string MHAParser::StrCnv::val2str (

const std::vector<int >&v)

Convert to string.

4.38.2.21 std::string MHAParser::StrCnv::val2str (

const std::vector<std::vector<int > > &v)

Convert to string.

4.38.2.22 std::string MHAParser::StrCnv::val2str (

const std::vector<std::string >&v)

Convert to string.

4.38.2.23 std::string MHAParser::StrCnv::val2str (

const std::vector<std::vector<ﬂoat > > &v)

Convert to string.

4.38.2.24 std::string MHAParser::StrCnv::val2str (

const std::vector<std::vector<mha_complex_t > > &v)

Convert to string.

4.39 MHAPlugin Namespace Reference

Namespace for openMHA plugin class templates and thread-safe runtime conﬁgurations.

Classes

• class cfg_chain_t

• class conﬁg_t

Template class for thread safe conﬁguration.

• class plugin_t

The template class for C++ openMHA plugins.

4.40 MHAPlugin_Resampling Namespace Reference 117

4.39.1 Detailed Description

Namespace for openMHA plugin class templates and thread-safe runtime conﬁgurations.

4.40 MHAPlugin_Resampling Namespace Reference

Classes

• class resampling_if_t

• class resampling_t

4.41 MHAPlugin_Split Namespace Reference

Classes

• class domain_handler_t

Handles domain-speciﬁc partial input and output signal.

• class dummy_threads_t

Dummy speciﬁcation of a thread platform: This class implements everything in a single thread.

• class posix_threads_t

Posix threads speciﬁcation of thread platform.

• class split_t

Implements split plugin.

• class splitted_part_t

The splitted_part_t (p. 751) instance manages the plugin that performs processing on the

reduced set of channels.

• class thread_platform_t

Basic interface for encapsulating thread creation, thread priority setting, and synchronization

on any threading platform (i.e., pthreads or win32threads).

• class uni_processor_t

An interface to a class that sports a process method with no parameters and no return value.

Enumerations

4.41.1 Detailed Description

A namespace for the split plugin. Helps testability and documentation.

118 CONTENTS

4.41.2 Enumeration Type Documentation

4.41.2.1 anonymous enum

Invalid thread priority.

Enumerator

INVALID_THREAD_PRIORITY

4.42 MHASignal Namespace Reference

Namespace for audio signal handling and processing classes.

Classes

• class async_rmslevel_t

Class for asynchronous level metering.

• class delay_spec_t

• class delay_t

Class to realize a simple delay of waveform streams.

• class delay_wave_t

Delayline containing wave fragments.

• class doublebuffer_t

Double-buffering class.

• class fft_t

• class hilbert_fftw_t

• class hilbert_t

Hilbert transformation of a waveform segment.

• class loop_wavefragment_t

Copy a ﬁxed waveform fragment to a series of waveform fragments of other size.

• class matrix_t

n-dimensional matrix with real or complex ﬂoating point values.

• class minphase_t

Minimal phase function.

• class quantizer_t

Simple simulation of ﬁxpoint quantization.

• class ringbuffer_t

A ringbuffer class for time domain audio signal, which makes no assumptions with respect to

fragment size.

• class schroeder_t

Schroeder tone complex class.

• class spectrum_t

a signal processing class for spectral data (based on mha_spec_t (p. 465))

4.42 MHASignal Namespace Reference 119

• class stat_t

• class subsample_delay_t

implements subsample delay in spectral domain.

• class uint_vector_t

Vector of unsigned values, used for size and index description of n-dimensional matrixes.

• class waveform_t

signal processing class for waveform data (based on mha_wave_t (p. 496))

Functions

• void for_each (mha_wave_t ∗s, mha_real_t(∗fun)(mha_real_t))

Apply a function to each element of a mha_wave_t (p. 496).

•mha_real_t lin2db (mha_real_t x)

Conversion from linear scale to dB (no SPL reference)

•mha_real_t db2lin (mha_real_t x)

Conversion from dB scale to linear (no SPL reference)

•mha_real_t pa2dbspl (mha_real_t x)

Conversion from linear Pascal scale to dB SPL.

•mha_real_t pa22dbspl (mha_real_t x, mha_real_t eps=1e-20f)

Conversion from squared Pascal scale to dB SPL.

•mha_real_t dbspl2pa (mha_real_t x)

Conversion from dB SPL to linear Pascal scale.

•mha_real_t smp2sec (mha_real_t n, mha_real_t srate)

conversion from samples to seconds

•mha_real_t sec2smp (mha_real_t sec, mha_real_t srate)

conversion from seconds to samples

•mha_real_t bin2freq (mha_real_t bin, unsigned fftlen, mha_real_t srate)

conversion from fft bin index to frequency

•mha_real_t freq2bin (mha_real_t freq, unsigned fftlen, mha_real_t srate)

conversion from frequency to fft bin index

•mha_real_t smp2rad (mha_real_t samples, unsigned bin, unsigned fftlen)

conversion from delay in samples to phase shift

•mha_real_t rad2smp (mha_real_t phase_shift, unsigned bin, unsigned fftlen)

conversion from phase shift to delay in samples

•template<class elem_type >

std::vector<elem_type >dupvec (std::vector<elem_type >vec, unsigned n)

Duplicate last vector element to match desired size.

•template<class elem_type >

std::vector<elem_type >dupvec_chk (std::vector<elem_type >vec, unsigned n)

Duplicate last vector element to match desired size, check for dimension.

• void copy_channel (mha_spec_t &self, const mha_spec_t &src, unsigned sch, un-

signed dch)

Copy one channel of a source signal.

• void copy_channel (mha_wave_t &self, const mha_wave_t &src, unsigned src_←-

channel, unsigned dest_channel)

120 CONTENTS

Copy one channel of a source signal.

•mha_real_t rmslevel (const mha_spec_t &s, unsigned int channel, unsigned int fftlen)

Return RMS level of a spectrum channel.

•mha_real_t colored_intensity (const mha_spec_t &s, unsigned int channel, unsigned

int fftlen, mha_real_t sqfreq_response[ ])

Colored spectrum intensity.

•mha_real_t maxabs (const mha_spec_t &s, unsigned int channel)

Find maximal absolute value.

•mha_real_t rmslevel (const mha_wave_t &s, unsigned int channel)

Return RMS level of a waveform channel.

•mha_real_t maxabs (const mha_wave_t &s, unsigned int channel)

Find maximal absolute value.

•mha_real_t maxabs (const mha_wave_t &s)

Find maximal absolute value.

•mha_real_t max (const mha_wave_t &s)

Find maximal value.

•mha_real_t min (const mha_wave_t &s)

Find minimal value.

•mha_real_t sumsqr_channel (const mha_wave_t &s, unsigned int channel)

Calculate sum of squared values in one channel.

•mha_real_t sumsqr_frame (const mha_wave_t &s, unsigned int frame)

Calculate sum over all channels of squared values.

• void scale (mha_spec_t ∗dest, const mha_wave_t ∗src)

• void limit (mha_wave_t &s, const mha_real_t &min, const mha_real_t &max)

Limit the singal in the waveform buffer to the range [min, max].

•template<class elem_type >

elem_type kth_smallest (elem_type array[ ], unsigned n, unsigned k)

Fast search for the kth smallest element of an array.

•template<class elem_type >

elem_type median (elem_type array[ ], unsigned n)

Fast median search.

•template<class elem_type >

elem_type mean (const std::vector<elem_type >&data, elem_type start_val)

Calculate average of elements in a vector.

•template<class elem_type >

std::vector<elem_type >quantile (std::vector<elem_type >data, const std::vector<

elem_type >&p)

Calculate quantile of elements in a vector.

• void saveas_mat4 (const mha_spec_t &data, const std::string &varname, FILE ∗fh)

Save a openMHA spectrum as a variable in a Matlab4 ﬁle.

• void saveas_mat4 (const mha_wave_t &data, const std::string &varname, FILE ∗fh)

Save a openMHA waveform as a variable in a Matlab4 ﬁle.

• void saveas_mat4 (const std::vector<mha_real_t >&data, const std::string &varname,

FILE ∗fh)

Save a ﬂoat vector as a variable in a Matlab4 ﬁle.

• void copy_permuted (mha_wave_t ∗dest, const mha_wave_t ∗src)

Copy contents of a waveform to a permuted waveform.

4.42 MHASignal Namespace Reference 121

Variables

• unsigned long int signal_counter = 0

Signal counter to produce signal ID strings.

4.42.1 Detailed Description

Namespace for audio signal handling and processing classes.

4.42.2 Function Documentation

4.42.2.1 void MHASignal::scale (

mha_spec_t ∗dest,

const mha_wave_t ∗src )

4.42.2.2 void MHASignal::limit (

mha_wave_t &s,

const mha_real_t &min,

const mha_real_t &max )

Limit the singal in the waveform buffer to the range [min, max].

Parameters

sThe signal to limit. The signal in this wave buffer is modiﬁed.

min lower limit

max upper limit

4.42.2.3 template<class elem_type >elem_type MHASignal::kth_smallest (

elem_type array[ ],

unsigned n,

unsigned k)

Fast search for the kth smallest element of an array.

The order of elements is altered, but not completely sorted. Using the algorithm from N. Wirth,

published in "Algorithms + data structures = programs", Prentice-Hall, 1976

Parameters

array Element array

122 CONTENTS

Postcondition

The order of elements in the array is altered. array[k] then holds the result.

Parameters

nnumber of elements in array

Precondition

n>= 1

Parameters

kThe k'th smalles element is returned: k = 0 returns the minimum, k = (n-1)/2 returns the

median, k=(n-1) returns the maximum

Precondition

k<n

Returns

The kth smallest array element

4.42.2.4 template<class elem_type >elem_type MHASignal::median (

elem_type array[ ],

unsigned n)[inline]

Fast median search.

The order of elements is altered, but not completely sorted.

Parameters

array Element array

Postcondition

The order of elements in the array is altered. array[(n-1)/2] then holds the median.

Parameters

nnumber of elements in array

4.42 MHASignal Namespace Reference 123

Precondition

n>= 1

Returns

The median of the array elements

4.42.2.5 template<class elem_type >elem_type MHASignal::mean (

const std::vector<elem_type >&data,

elem_type start_val )[inline]

Calculate average of elements in a vector.

Parameters

data Input vector

start_val Value for initialization of the return value before sum.

Returns

The average of the vector elements

4.42.2.6 template<class elem_type >std::vector<elem_type>MHASignal::quantile (

std::vector<elem_type >data,

const std::vector<elem_type >&p)[inline]

Calculate quantile of elements in a vector.

Parameters

data Input vector

pVector of probability values.

Returns

Vector of quantiles of input data, one entry for each probability value.

4.42.2.7 void MHASignal::saveas_mat4 (

const mha_spec_t &data,

const std::string & varname,

FILE ∗fh )

Save a openMHA spectrum as a variable in a Matlab4 ﬁle.

124 CONTENTS

Parameters

data openMHA spectrum to be saved.

varname Matlab variable name (Matlab4 limitations on maximal length are not checked).

fh File handle to Matlab4 ﬁle.

4.42.2.8 void MHASignal::saveas_mat4 (

const mha_wave_t &data,

const std::string & varname,

FILE ∗fh )

Save a openMHA waveform as a variable in a Matlab4 ﬁle.

Parameters

data openMHA waveform to be saved.

varname Matlab variable name (Matlab4 limitations on maximal length are not checked).

fh File handle to Matlab4 ﬁle.

4.42.2.9 void MHASignal::saveas_mat4 (

const std::vector<mha_real_t >&data,

const std::string & varname,

FILE ∗fh )

Save a ﬂoat vector as a variable in a Matlab4 ﬁle.

Parameters

data Float vector to be saved.

varname Matlab variable name (Matlab4 limitations on maximal length are not checked).

fh File handle to Matlab4 ﬁle.

4.42.2.10 void MHASignal::copy_permuted (

mha_wave_t ∗dest,

const mha_wave_t ∗src )

Copy contents of a waveform to a permuted waveform.

Parameters

dest Destination waveform

src Source waveform

The total size of src and dest must be the same, num_frames and num_channels must be

4.43 MHASndFile Namespace Reference 125

exchanged in dest.

4.42.3 Variable Documentation

4.42.3.1 unsigned long int MHASignal::signal_counter = 0

Signal counter to produce signal ID strings.

4.43 MHASndFile Namespace Reference

Classes

• class sf_t

• class sf_wave_t

4.44 MHATableLookup Namespace Reference

Namespace for table lookup classes.

Classes

• class linear_table_t

Class for interpolation with equidistant x values.

• class table_t

• class xy_table_t

Class for interpolation with non-equidistant x values.

4.44.1 Detailed Description

Namespace for table lookup classes.

4.45 MHAWindow Namespace Reference

Collection of Window types.

126 CONTENTS

Classes

• class bartlett_t

Bartlett window.

• class base_t

Common base for window types.

• class blackman_t

Blackman window.

• class fun_t

Generic window based on a generator function.

• class hamming_t

Hamming window.

• class hanning_t

von-Hann window

• class rect_t

Rectangular window.

• class user_t

User deﬁned window.

Functions

• ﬂoat rect (ﬂoat)

Rectangular window function.

• ﬂoat bartlett (ﬂoat)

Bartlett window function.

• ﬂoat hanning (ﬂoat)

Hanning window function.

• ﬂoat hamming (ﬂoat)

Hamming window function.

• ﬂoat blackman (ﬂoat)

Blackman window function.

4.45.1 Detailed Description

Collection of Window types.

4.45.2 Function Documentation

4.45.2.1 ﬂoat MHAWindow::rect (

ﬂoat x)

Rectangular window function.

4.46 multibandcompressor Namespace Reference 127

4.45.2.2 ﬂoat MHAWindow::bartlett (

ﬂoat x)

Bartlett window function.

4.45.2.3 ﬂoat MHAWindow::hanning (

ﬂoat x)

Hanning window function.

4.45.2.4 ﬂoat MHAWindow::hamming (

ﬂoat x)

Hamming window function.

4.45.2.5 ﬂoat MHAWindow::blackman (

ﬂoat x)

Blackman window function.

4.46 multibandcompressor Namespace Reference

Classes

• class fftfb_plug_t

• class interface_t

• class plugin_signals_t

4.47 noisePowProposedScale Namespace Reference

Classes

• class interface_t

• class noisePowProposed

4.48 overlapadd Namespace Reference

Classes

• class overlapadd_if_t

• class overlapadd_t

128 CONTENTS

4.49 plingploing Namespace Reference

All classes for the plingploing music generator live in this namespace.

Classes

• class if_t

Plugin class of the plingploing music generator.

• class plingploing_t

Run-time conﬁguration of the plingploing music generator.

Functions

• double drand (double a, double b)

4.49.1 Detailed Description

All classes for the plingploing music generator live in this namespace.

4.49.2 Function Documentation

4.49.2.1 double plingploing::drand (

double a,

double b)

4.50 PluginLoader Namespace Reference

Classes

• class conﬁg_ﬁle_splitter_t

• class fourway_processor_t

This abstract class deﬁnes the interface for classes that implement all types of signal domain

processing supported by the MHA: wave2wave, spec2spec, wave2spec, and spec2wave.

• class mhapluginloader_t

Functions

• const char ∗mhastrdomain (mha_domain_t)

• void mhaconﬁg_compare (const mhaconﬁg_t &req, const mhaconﬁg_t &avail, const

std::string &pref="")

Compare two mhaconﬁg_t (p. 504) structures, and report differences as an error.

4.51 route Namespace Reference 129

4.50.1 Function Documentation

4.50.1.1 const char ∗PluginLoader::mhastrdomain (

mha_domain_t d)

4.50.1.2 void PluginLoader::mhaconﬁg_compare (

const mhaconﬁg_t &req,

const mhaconﬁg_t &avail,

const std::string & pref = "" )

Compare two mhaconﬁg_t (p. 504) structures, and report differences as an error.

Parameters

req Expected mhaconﬁg_t (p. 504) structure

avail Available mhaconﬁg_t (p. 504) structure

pref Preﬁx for error messages

4.51 route Namespace Reference

Classes

• class interface_t

• class process_t

4.52 shadowﬁlter_begin Namespace Reference

Classes

• class cfg_t

• class shadowﬁlter_begin_t

4.53 shadowﬁlter_end Namespace Reference

Classes

• class cfg_t

• class shadowﬁlter_end_t

130 CONTENTS

4.54 smoothgains_bridge Namespace Reference

Classes

• class overlapadd_if_t

• class smoothspec_wrap_t

4.55 testplugin Namespace Reference

Classes

• class ac_parser_t

• class conﬁg_parser_t

• class if_t

• class signal_parser_t

5 Class Documentation

5.1 ac2lsl::ac2lsl_t Class Reference

Plugin class of ac2lsl (p. 78).

Inheritance diagram for ac2lsl::ac2lsl_t:

ac2lsl::ac2lsl_t

MHAPlugin::plugin_t

< cfg_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< cfg_t >

5.1 ac2lsl::ac2lsl_t Class Reference 131

Public Member Functions

•ac2lsl_t (algo_comm_t iac, const char ∗chain, const char ∗algo)

• void prepare (mhaconﬁg_t &)

Prepare constructs the vector of bridge variables and locks the conﬁguration, then calls up-

date() (p. 133).

•mha_wave_t ∗process (mha_wave_t ∗s)

Processing fct for waveforms.

•mha_spec_t ∗process (mha_spec_t ∗s)

Processing fct for spectra.

• void process ()

Process function.

• void release ()

Release fct.

Private Member Functions

• std::vector<std::string >get_all_names_from_ac_space (const algo_comm_t &ac)

const

Retrieves all variable names from the AC space.

• void update ()

Construct new runtime conﬁguration.

Private Attributes

•MHAParser::vstring_t vars

•MHAParser::string_t source_id

•MHAParser::bool_t rt_strict

•MHAParser::bool_t activate

•MHAParser::int_t skip

•MHAEvents::patchbay_t<ac2lsl_t >patchbay

• bool is_ﬁrst_run

Additional Inherited Members

5.1.1 Detailed Description

Plugin class of ac2lsl (p. 78).

132 CONTENTS

5.1.2 Constructor & Destructor Documentation

5.1.2.1 ac2lsl::ac2lsl_t::ac2lsl_t (

algo_comm_t iac,

const char ∗chain,

const char ∗algo )

5.1.3 Member Function Documentation

5.1.3.1 void ac2lsl::ac2lsl_t::prepare (

mhaconﬁg_t &cf )[virtual]

Prepare constructs the vector of bridge variables and locks the conﬁguration, then calls up-

date() (p. 133).

Implements MHAPlugin::plugin_t<cfg_t >(p. 730).

5.1.3.2 mha_wave_t∗ac2lsl::ac2lsl_t::process (

mha_wave_t ∗s)[inline]

Processing fct for waveforms.

Calls process(void) (p. 132).

5.1.3.3 mha_spec_t∗ac2lsl::ac2lsl_t::process (

mha_spec_t ∗s)[inline]

Processing fct for spectra.

Calls process(void) (p. 132).

5.1.3.4 void ac2lsl::ac2lsl_t::process (

void )

Process function.

Checks once if the plugin is run in a real-time thread and throws if rt_strict is true, then forwards

to cfg_t::process() (p. 135).

5.1.3.5 void ac2lsl::ac2lsl_t::release ( ) [virtual]

Release fct.

Unlocks variable name list

Reimplemented from MHAPlugin::plugin_t<cfg_t >(p. 731).

5.1.3.6 std::vector<std::string >ac2lsl::ac2lsl_t::get_all_names_from_ac_space (

const algo_comm_t &ac ) const [private]

Retrieves all variable names from the AC space.

5.2 ac2lsl::cfg_t Class Reference 133

Parameters

ac AC space

Returns

Vector of variable names

5.1.3.7 void ac2lsl::ac2lsl_t::update ( ) [private]

Construct new runtime conﬁguration.

5.1.4 Member Data Documentation

5.1.4.1 MHAParser::vstring_t ac2lsl::ac2lsl_t::vars [private]

5.1.4.2 MHAParser::string_t ac2lsl::ac2lsl_t::source_id [private]

5.1.4.3 MHAParser::bool_t ac2lsl::ac2lsl_t::rt_strict [private]

5.1.4.4 MHAParser::bool_t ac2lsl::ac2lsl_t::activate [private]

5.1.4.5 MHAParser::int_t ac2lsl::ac2lsl_t::skip [private]

5.1.4.6 MHAEvents::patchbay_t<ac2lsl_t>ac2lsl::ac2lsl_t::patchbay [private]

5.1.4.7 bool ac2lsl::ac2lsl_t::is_ﬁrst_run [private]

The documentation for this class was generated from the following ﬁle:

•ac2lsl.cpp

5.2 ac2lsl::cfg_t Class Reference

Runtime conﬁguration class of the ac2lsl (p. 78) plugin.

Public Member Functions

•cfg_t (const algo_comm_t &ac_, unsigned skip_, const std::string &source_id, const

std::vector<std::string >&varnames_, double rate)

C'tor of ac2lsl (p. 78) run time conﬁguration.

• void process ()

134 CONTENTS

Private Member Functions

• void create_or_replace_var (const std::string &name, const comm_var_t &v)

• void check_vars ()

• void update_varlist ()

Private Attributes

• std::map<std::string, std::unique_ptr<save_var_base_t > > varlist

Maps variable name to unique ptr's of ac to lsl bridges.

• unsigned skipcnt

Counter of frames to skip.

• const unsigned skip

Number of frames to skip after each send.

• const double srate

Sampling rate of the stream.

• const std::string source_id

User conﬁgurable source id.

• const algo_comm_t &ac

Handle to the ac space.

5.2.1 Detailed Description

Runtime conﬁguration class of the ac2lsl (p. 78) plugin.

5.2.2 Constructor & Destructor Documentation

5.2.2.1 cfg_t::cfg_t (

const algo_comm_t &ac_,

unsigned skip_,

const std::string & source_id,

const std::vector<std::string >&varnames_,

double rate )

C'tor of ac2lsl (p. 78) run time conﬁguration.

Parameters

ac_ AC space, source of data to send over LSL

skip_ Number of frames to skip after each send

source_←-

id_

LSL identiﬁer for this data stream

varnames←-

Names of AC variables to send over LSL

rate Rate with wich chunks of data are sent to the LSL stream. Usually the rate

with which process calls happen, but may be lower due to the subsampling

caused by skip_

5.2 ac2lsl::cfg_t Class Reference 135

5.2.3 Member Function Documentation

5.2.3.1 void cfg_t::create_or_replace_var (

const std::string & name,

const comm_var_t &v)[private]

5.2.3.2 void ac2lsl::cfg_t::check_vars ( ) [private]

5.2.3.3 void cfg_t::update_varlist ( ) [private]

5.2.3.4 void cfg_t::process (

void )

5.2.4 Member Data Documentation

5.2.4.1 std::map<std::string, std::unique_ptr<save_var_base_t> > ac2lsl::cfg_t::varlist

[private]

Maps variable name to unique ptr's of ac to lsl bridges.

5.2.4.2 unsigned ac2lsl::cfg_t::skipcnt [private]

Counter of frames to skip.

5.2.4.3 const unsigned ac2lsl::cfg_t::skip [private]

Number of frames to skip after each send.

5.2.4.4 const double ac2lsl::cfg_t::srate [private]

Sampling rate of the stream.

5.2.4.5 const std::string ac2lsl::cfg_t::source_id [private]

User conﬁgurable source id.

5.2.4.6 const algo_comm_t& ac2lsl::cfg_t::ac [private]

Handle to the ac space.

The documentation for this class was generated from the following ﬁle:

•ac2lsl.cpp

136 CONTENTS

5.3 ac2lsl::save_var_base_t Class Reference

Interface for ac to lsl bridge variable.

Inheritance diagram for ac2lsl::save_var_base_t:

ac2lsl::save_var_base_t

ac2lsl::save_var_t< T > ac2lsl::save_var_t

< mha_complex_t >

Public Member Functions

• virtual void send_frame ()=0

• virtual void ∗get_buf_address () const noexcept=0

• virtual void set_buf_address (void ∗data)=0

• virtual lsl::stream_info info () const noexcept=0

• virtual unsigned data_type () const noexcept=0

• virtual unsigned num_entries () const noexcept=0

• virtual ∼save_var_base_t ()=default

5.3.1 Detailed Description

Interface for ac to lsl bridge variable.

5.3.2 Constructor & Destructor Documentation

5.3.2.1 virtual ac2lsl::save_var_base_t::∼save_var_base_t ( ) [virtual],[default]

5.3.3 Member Function Documentation

5.3.3.1 virtual void ac2lsl::save_var_base_t::send_frame ( ) [pure virtual]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 143), and ac2lsl::save_var_t<

T>(p. 140).

5.4 ac2lsl::save_var_t<T>Class Template Reference 137

5.3.3.2 virtual void∗ac2lsl::save_var_base_t::get_buf_address ( ) const [pure virtual],

[noexcept]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 142), and ac2lsl::save_var_t<

T>(p. 139).

5.3.3.3 virtual void ac2lsl::save_var_base_t::set_buf_address (

void ∗data )[pure virtual]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 142), and ac2lsl::save_var_t<

T>(p. 139).

5.3.3.4 virtual lsl::stream_info ac2lsl::save_var_base_t::info ( ) const [pure virtual],

[noexcept]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 142), and ac2lsl::save_var_t<

T>(p. 139).

5.3.3.5 virtual unsigned ac2lsl::save_var_base_t::data_type ( ) const [pure virtual],

[noexcept]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 142), and ac2lsl::save_var_t<

T>(p. 140).

5.3.3.6 virtual unsigned ac2lsl::save_var_base_t::num_entries ( ) const [pure virtual],

[noexcept]

Implemented in ac2lsl::save_var_t<mha_complex_t >(p. 142), and ac2lsl::save_var_t<

T>(p. 139).

The documentation for this class was generated from the following ﬁle:

•ac2lsl.cpp

5.4 ac2lsl::save_var_t<T>Class Template Reference

Implementation for all ac to lsl bridges except complex types.

Inheritance diagram for ac2lsl::save_var_t<T>:

ac2lsl::save_var_t< T >

ac2lsl::save_var_base_t

138 CONTENTS

Public Member Functions

•save_var_t (const std::string &name_, const std::string &type_, unsigned num_entries_,

const mha_real_t rate_, const lsl::channel_format_t format_, const std::string &source←-

_id_, void ∗data_, const unsigned data_type_)

C'tor of generic ac to lsl bridge.

• virtual void ∗get_buf_address () const noexceptoverride

Get buffer address as void pointer.

• virtual void set_buf_address (void ∗data) override

Cast the input pointer to the appropriate type and set the buffer address.

• virtual lsl::stream_info info () const noexceptoverride

Get stream info object from stream outlet.

• virtual unsigned num_entries () const noexceptoverride

Get number of entries in the stream object.

• virtual unsigned data_type () const noexceptoverride

Get data type id according MHA convention.

• virtual ∼save_var_t ()=default

• virtual void send_frame () override

Send a frame to lsl.

Private Attributes

• lsl::stream_outlet stream

LSL stream outlet.

• T ∗buf

Pointer to data buffer of the ac variable.

• const unsigned data_type_

Data type id according to MHA convention.

5.4.1 Detailed Description

template<typename T>

class ac2lsl::save_var_t<T>

Implementation for all ac to lsl bridges except complex types.

5.4.2 Constructor & Destructor Documentation

5.4.2.1 template<typename T >ac2lsl::save_var_t<T>::save_var_t (

const std::string & name_,

const std::string & type_,

unsigned num_entries_,

const mha_real_t rate_,

const lsl::channel_format_t format_,

const std::string & source_id_,

void ∗data_,

const unsigned data_type_ )[inline]

C'tor of generic ac to lsl bridge.

5.4 ac2lsl::save_var_t<T>Class Template Reference 139

Parameters

info LSL stream info object containing metadata

data Pointer to data buffer of the ac variable

data_type Type id of the stream, in mha convention. Should be set to one if not a vector.

5.4.2.2 template<typename T >virtual ac2lsl::save_var_t<T>::∼save_var_t ( ) [virtual],

[default]

5.4.3 Member Function Documentation

5.4.3.1 template<typename T >virtual void∗ac2lsl::save_var_t<T>::get_buf_address ( ) const

[inline],[override],[virtual],[noexcept]

Get buffer address as void pointer.

Returns

Adress of the data buffer

Implements ac2lsl::save_var_base_t (p. 137).

5.4.3.2 template<typename T >virtual void ac2lsl::save_var_t<T>::set_buf_address (

void ∗data )[inline],[override],[virtual]

Cast the input pointer to the appropriate type and set the buffer address.

Parameters

data New buffer address

Implements ac2lsl::save_var_base_t (p. 137).

5.4.3.3 template<typename T >virtual lsl::stream_info ac2lsl::save_var_t<T>::info ( ) const

[inline],[override],[virtual],[noexcept]

Get stream info object from stream outlet.

Implements ac2lsl::save_var_base_t (p. 137).

5.4.3.4 template<typename T >virtual unsigned ac2lsl::save_var_t<T>::num_entries ( ) const

[inline],[override],[virtual],[noexcept]

Get number of entries in the stream object.

Implements ac2lsl::save_var_base_t (p. 137).

140 CONTENTS

5.4.3.5 template<typename T >virtual unsigned ac2lsl::save_var_t<T>::data_type ( ) const

[inline],[override],[virtual],[noexcept]

Get data type id according MHA convention.

Implements ac2lsl::save_var_base_t (p. 137).

5.4.3.6 template<typename T >virtual void ac2lsl::save_var_t<T>::send_frame ( ) [inline],

[override],[virtual]

Send a frame to lsl.

Implements ac2lsl::save_var_base_t (p. 136).

5.4.4 Member Data Documentation

5.4.4.1 template<typename T >lsl::stream_outlet ac2lsl::save_var_t<T>::stream [private]

LSL stream outlet.

Interface to lsl

5.4.4.2 template<typename T >T∗ac2lsl::save_var_t<T>::buf [private]

Pointer to data buffer of the ac variable.

5.4.4.3 template<typename T >const unsigned ac2lsl::save_var_t<T>::data_type_

[private]

Data type id according to MHA convention.

The documentation for this class was generated from the following ﬁle:

•ac2lsl.cpp

5.5 ac2lsl::save_var_t<mha_complex_t >Class Template Reference

Template specialization of the ac2lsl (p. 78) bridge to take care of complex numbers.

Inheritance diagram for ac2lsl::save_var_t<mha_complex_t >:

ac2lsl::save_var_t

< mha_complex_t >

ac2lsl::save_var_base_t

5.5 ac2lsl::save_var_t<mha_complex_t >Class Template

Reference 141

Public Member Functions

•save_var_t (const std::string &name_, const std::string &type_, const unsigned num_←-

entries_, const mha_real_t rate_, const lsl::channel_format_t format_, const std::string

&source_id_, void ∗data_)

C'tor of specialization for complex types.

• virtual void ∗get_buf_address () const noexceptoverride

• virtual void set_buf_address (void ∗data) override

• virtual lsl::stream_info info () const noexceptoverride

Get buffer address as void pointer.

• virtual unsigned num_entries () const noexceptoverride

Get number of entries in the stream object.

• virtual unsigned data_type () const noexceptoverride

Cast the input pointer to the appropriate type and set the buffer address.

• virtual ∼save_var_t ()=default

• virtual void send_frame () override

Send a frame of complex types.

Private Attributes

• lsl::stream_outlet stream

LSL stream outlet.

•mha_complex_t ∗buf

Pointer to data buffer of the ac variable.

5.5.1 Detailed Description

template<>

class ac2lsl::save_var_t<mha_complex_t >

Template specialization of the ac2lsl (p. 78) bridge to take care of complex numbers.

This specialization is needed because lsl does not support complex numbers. Order is [re(0),

im(0), re(1), im(1), ....]

5.5.2 Constructor & Destructor Documentation

5.5.2.1 ac2lsl::save_var_t<mha_complex_t >::save_var_t (

const std::string & name_,

const std::string & type_,

const unsigned num_entries_,

const mha_real_t rate_,

const lsl::channel_format_t format_,

const std::string & source_id_,

void ∗data_ )[inline]

C'tor of specialization for complex types.

See generic c'tor for details.

142 CONTENTS

5.5.2.2 virtual ac2lsl::save_var_t<mha_complex_t >::∼save_var_t ( ) [virtual],

[default]

5.5.3 Member Function Documentation

5.5.3.1 virtual void∗ac2lsl::save_var_t<mha_complex_t >::get_buf_address ( ) const

[inline],[override],[virtual],[noexcept]

Implements ac2lsl::save_var_base_t (p. 137).

5.5.3.2 virtual void ac2lsl::save_var_t<mha_complex_t >::set_buf_address (

void ∗data )[inline],[override],[virtual]

Implements ac2lsl::save_var_base_t (p. 137).

5.5.3.3 virtual lsl::stream_info ac2lsl::save_var_t<mha_complex_t >::info ( ) const

[inline],[override],[virtual],[noexcept]

Get buffer address as void pointer.

Returns

Adress of the data buffer

Implements ac2lsl::save_var_base_t (p. 137).

5.5.3.4 virtual unsigned ac2lsl::save_var_t<mha_complex_t >::num_entries ( ) const

[inline],[override],[virtual],[noexcept]

Get number of entries in the stream object.

Implements ac2lsl::save_var_base_t (p. 137).

5.5.3.5 virtual unsigned ac2lsl::save_var_t<mha_complex_t >::data_type ( ) const

[inline],[override],[virtual],[noexcept]

Cast the input pointer to the appropriate type and set the buffer address.

Parameters

data New buffer address

Implements ac2lsl::save_var_base_t (p. 137).

5.6 ac2lsl::type_info Struct Reference 143

5.5.3.6 virtual void ac2lsl::save_var_t<mha_complex_t >::send_frame ( ) [inline],

[override],[virtual]

Send a frame of complex types.

Complex numbers are stored as alternating real and imaginary parts. An array of complex

numbers in memory can be reinterpreted as a vector of real numbers that correspond to real

and imaginary parts. LSL does not support complex types directly. Send one vector containing

{buf[0].re,buf[0].im,buf[1].re,buf[1].im,...} instead.

Implements ac2lsl::save_var_base_t (p. 136).

5.5.4 Member Data Documentation

5.5.4.1 lsl::stream_outlet ac2lsl::save_var_t<mha_complex_t >::stream [private]

LSL stream outlet.

Interface to lsl

5.5.4.2 mha_complex_t∗ac2lsl::save_var_t<mha_complex_t >::buf [private]

Pointer to data buffer of the ac variable.

The documentation for this class was generated from the following ﬁle:

•ac2lsl.cpp

5.6 ac2lsl::type_info Struct Reference

Public Attributes

• const std::string name

• const lsl::channel_format_t format

5.6.1 Member Data Documentation

5.6.1.1 const std::string ac2lsl::type_info::name

5.6.1.2 const lsl::channel_format_t ac2lsl::type_info::format

The documentation for this struct was generated from the following ﬁle:

•ac2lsl.cpp

144 CONTENTS

5.7 ac2osc_t Class Reference

Plugin class of the ac2osc plugin.

Inheritance diagram for ac2osc_t:

ac2osc_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•ac2osc_t (algo_comm_t iac, const char ∗chain, const char ∗algo)

C'tor of plugin class.

• void prepare (mhaconﬁg_t &)

•mha_wave_t ∗process (mha_wave_t ∗s)

Processing fct for waveforms.

•mha_spec_t ∗process (mha_spec_t ∗s)

Processing fct for spectra.

• void process ()

Process function.

• void release ()

Release frees osc related memory, does cleanup.

Private Member Functions

• void send_osc_ﬂoat ()

• void update_mode ()

Start/Stop sending of messages.

5.7 ac2osc_t Class Reference 145

Private Attributes

•MHAParser::string_t host

OSC server host name.

•MHAParser::string_t port

OSC server port.

•MHAParser::int_t ttl

Time-to-live of UDP packages.

•MHAParser::vstring_t vars

List of AC variables to be saved, empty for all.

•MHAParser::kw_t mode

Record mode.

•MHAParser::int_t skip

number of frames to skip after sending

•MHAParser::bool_t rt_strict

abort if used in real-time thread?

•MHA_AC::acspace2matrix_t ∗acspace

•MHAEvents::patchbay_t<ac2osc_t >patchbay

• bool b_record

• uint8_t ∗rtmem

• ﬂoat framerate

• int skipcnt

• lo_address lo_addr

• bool is_ﬁrst_run

Additional Inherited Members

5.7.1 Detailed Description

Plugin class of the ac2osc plugin.

5.7.2 Constructor & Destructor Documentation

5.7.2.1 ac2osc_t::ac2osc_t (

algo_comm_t iac,

const char ∗chain,

const char ∗algo )

C'tor of plugin class.

146 CONTENTS

5.7.3 Member Function Documentation

5.7.3.1 void ac2osc_t::prepare (

mhaconﬁg_t &cf )[virtual]

Implements MHAPlugin::plugin_t<int >(p. 730).

5.7.3.2 mha_wave_t∗ac2osc_t::process (

mha_wave_t ∗s)[inline]

Processing fct for waveforms.

Calls process(void) (p. 146).

5.7.3.3 mha_spec_t∗ac2osc_t::process (

mha_spec_t ∗s)[inline]

Processing fct for spectra.

Calls process(void) (p. 146).

5.7.3.4 void ac2osc_t::process (

void )

Process function.

Checks once if the plugin is run in a real-time thread and throws if rt_strict is true, sends osc

messages according to conﬁg.

5.7.3.5 void ac2osc_t::release (

void ) [virtual]

Release frees osc related memory, does cleanup.

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.7.3.6 void ac2osc_t::send_osc_ﬂoat ( ) [private]

5.7.3.7 void ac2osc_t::update_mode ( ) [private]

Start/Stop sending of messages.

5.7.4 Member Data Documentation

5.7.4.1 MHAParser::string_t ac2osc_t::host [private]

OSC server host name.

5.7 ac2osc_t Class Reference 147

5.7.4.2 MHAParser::string_t ac2osc_t::port [private]

OSC server port.

5.7.4.3 MHAParser::int_t ac2osc_t::ttl [private]

Time-to-live of UDP packages.

5.7.4.4 MHAParser::vstring_t ac2osc_t::vars [private]

List of AC variables to be saved, empty for all.

5.7.4.5 MHAParser::kw_t ac2osc_t::mode [private]

Record mode.

5.7.4.6 MHAParser::int_t ac2osc_t::skip [private]

number of frames to skip after sending

5.7.4.7 MHAParser::bool_t ac2osc_t::rt_strict [private]

abort if used in real-time thread?

5.7.4.8 MHA_AC::acspace2matrix_t∗ac2osc_t::acspace [private]

5.7.4.9 MHAEvents::patchbay_t<ac2osc_t>ac2osc_t::patchbay [private]

5.7.4.10 bool ac2osc_t::b_record [private]

5.7.4.11 uint8_t∗ac2osc_t::rtmem [private]

5.7.4.12 ﬂoat ac2osc_t::framerate [private]

5.7.4.13 int ac2osc_t::skipcnt [private]

5.7.4.14 lo_address ac2osc_t::lo_addr [private]

5.7.4.15 bool ac2osc_t::is_ﬁrst_run [private]

The documentation for this class was generated from the following ﬁle:

•ac2osc.cpp

148 CONTENTS

5.8 ac2wave_if_t Class Reference

Inheritance diagram for ac2wave_if_t:

ac2wave_if_t

MHAPlugin::plugin_t

< ac2wave_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< ac2wave_t >

Public Member Functions

•ac2wave_if_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_wave_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void update ()

Private Attributes

•MHAParser::string_t name

•MHAParser::ﬂoat_t gain_in

•MHAParser::ﬂoat_t gain_ac

•MHAParser::int_t delay_in

•MHAParser::int_t delay_ac

•MHASignal::waveform_t ∗zeros

• bool prepared

•MHAEvents::patchbay_t<ac2wave_if_t >patchbay

5.8 ac2wave_if_t Class Reference 149

Additional Inherited Members

5.8.1 Constructor & Destructor Documentation

5.8.1.1 ac2wave_if_t::ac2wave_if_t (

const algo_comm_t &iac,

const std::string & ith,

const std::string & ial )

5.8.2 Member Function Documentation

5.8.2.1 mha_wave_t ∗ac2wave_if_t::process (

mha_spec_t ∗)

5.8.2.2 mha_wave_t ∗ac2wave_if_t::process (

mha_wave_t ∗s)

5.8.2.3 void ac2wave_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<ac2wave_t >(p. 730).

5.8.2.4 void ac2wave_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<ac2wave_t >(p. 731).

5.8.2.5 void ac2wave_if_t::update ( ) [private]

5.8.3 Member Data Documentation

5.8.3.1 MHAParser::string_t ac2wave_if_t::name [private]

5.8.3.2 MHAParser::ﬂoat_t ac2wave_if_t::gain_in [private]

5.8.3.3 MHAParser::ﬂoat_t ac2wave_if_t::gain_ac [private]

5.8.3.4 MHAParser::int_t ac2wave_if_t::delay_in [private]

5.8.3.5 MHAParser::int_t ac2wave_if_t::delay_ac [private]

5.8.3.6 MHASignal::waveform_t∗ac2wave_if_t::zeros [private]

5.8.3.7 bool ac2wave_if_t::prepared [private]

5.8.3.8 MHAEvents::patchbay_t<ac2wave_if_t>ac2wave_if_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•ac2wave.cpp

150 CONTENTS

5.9 ac2wave_t Class Reference

Public Member Functions

•ac2wave_t (unsigned int frames_, unsigned int channels_, algo_comm_t ac_, std::string

name_, ﬂoat gain_in_, ﬂoat gain_ac_, unsigned int delay_in_, unsigned int delay_ac_)

•mha_wave_t ∗process (mha_wave_t ∗)

Private Attributes

• unsigned int frames

• unsigned int channels

•mha_wave_t w

•algo_comm_t ac

• std::string name

•MHASignal::delay_wave_t delay_in

•MHASignal::delay_wave_t delay_ac

•mha_real_t gain_in

•mha_real_t gain_ac

5.9.1 Constructor & Destructor Documentation

5.9.1.1 ac2wave_t::ac2wave_t (

unsigned int frames_,

unsigned int channels_,

algo_comm_t ac_,

std::string name_,

ﬂoat gain_in_,

ﬂoat gain_ac_,

unsigned int delay_in_,

unsigned int delay_ac_ )

5.9.2 Member Function Documentation

5.9.2.1 mha_wave_t ∗ac2wave_t::process (

mha_wave_t ∗s)

5.9.3 Member Data Documentation

5.9.3.1 unsigned int ac2wave_t::frames [private]

5.9.3.2 unsigned int ac2wave_t::channels [private]

5.9.3.3 mha_wave_t ac2wave_t::w [private]

5.10 acConcat_wave Class Reference 151

5.9.3.4 algo_comm_t ac2wave_t::ac [private]

5.9.3.5 std::string ac2wave_t::name [private]

5.9.3.6 MHASignal::delay_wave_t ac2wave_t::delay_in [private]

5.9.3.7 MHASignal::delay_wave_t ac2wave_t::delay_ac [private]

5.9.3.8 mha_real_t ac2wave_t::gain_in [private]

5.9.3.9 mha_real_t ac2wave_t::gain_ac [private]

The documentation for this class was generated from the following ﬁle:

•ac2wave.cpp

5.10 acConcat_wave Class Reference

Inheritance diagram for acConcat_wave:

acConcat_wave

MHAPlugin::plugin_t

< acConcat_wave_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< acConcat_wave_config >

152 CONTENTS

Public Member Functions

•acConcat_wave (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

Constructs our plugin.

•∼acConcat_wave ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::int_t num_AC

•MHAParser::string_t preﬁx_names_AC

•MHAParser::vint_t samples_AC

•MHAParser::string_t name_con_AC

•MHAParser::int_t numchannels

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<acConcat_wave >patchbay

Additional Inherited Members

5.10.1 Constructor & Destructor Documentation

5.10.1.1 acConcat_wave::acConcat_wave (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.10 acConcat_wave Class Reference 153

5.10.1.2 acConcat_wave::∼acConcat_wave ( )

5.10.2 Member Function Documentation

5.10.2.1 mha_wave_t ∗acConcat_wave::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.10.2.2 void acConcat_wave::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<acConcat_wave_conﬁg >(p. 730).

5.10.2.3 void acConcat_wave::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<acConcat_wave_conﬁg >(p. 731).

5.10.2.4 void acConcat_wave::update_cfg ( ) [private]

5.10.3 Member Data Documentation

5.10.3.1 MHAParser::int_t acConcat_wave::num_AC

5.10.3.2 MHAParser::string_t acConcat_wave::preﬁx_names_AC

5.10.3.3 MHAParser::vint_t acConcat_wave::samples_AC

5.10.3.4 MHAParser::string_t acConcat_wave::name_con_AC

5.10.3.5 MHAParser::int_t acConcat_wave::numchannels

5.10.3.6 MHAEvents::patchbay_t<acConcat_wave>acConcat_wave::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•acConcat_wave.h

•acConcat_wave.cpp

154 CONTENTS

5.11 acConcat_wave_conﬁg Class Reference

Public Member Functions

•acConcat_wave_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg, acConcat_←-

wave ∗_concat)

•∼acConcat_wave_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

Public Attributes

•algo_comm_t &ac

• std::vector<std::string >strNames_AC

• std::vector<int >numSamples_AC

•mha_wave_t vGCC

•MHA_AC::waveform_t ∗vGCC_con

5.11.1 Constructor & Destructor Documentation

5.11.1.1 acConcat_wave_conﬁg::acConcat_wave_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

acConcat_wave ∗_concat )

5.11.1.2 acConcat_wave_conﬁg::∼acConcat_wave_conﬁg ( )

5.11.2 Member Function Documentation

5.11.2.1 mha_wave_t ∗acConcat_wave_conﬁg::process (

mha_wave_t ∗wave )

5.11.3 Member Data Documentation

5.11.3.1 algo_comm_t& acConcat_wave_conﬁg::ac

5.11.3.2 std::vector<std::string>acConcat_wave_conﬁg::strNames_AC

5.11.3.3 std::vector<int>acConcat_wave_conﬁg::numSamples_AC

5.11.3.4 mha_wave_t acConcat_wave_conﬁg::vGCC

5.11.3.5 MHA_AC::waveform_t∗acConcat_wave_conﬁg::vGCC_con

The documentation for this class was generated from the following ﬁles:

•acConcat_wave.h

•acConcat_wave.cpp

5.12 acmon::ac_monitor_t Class Reference 155

5.12 acmon::ac_monitor_t Class Reference

A class for converting AC variables to Parser monitors of correct type.

Public Member Functions

•ac_monitor_t (MHAParser::parser_t &parent, const std::string &name_, algo_comm←-

_t ac, bool use_matrix)

Converts AC variable to parser monitor.

• void getvar (algo_comm_t ac)

Update values of monitor.

Public Attributes

• std::string name

name of AC variable and parser monitor

• std::string dimstr

columns x rows

•MHAParser::vﬂoat_mon_t mon

Monitor used for real vectors.

•MHAParser::mﬂoat_mon_t mon_mat

Monitor used for real matrices.

•MHAParser::vcomplex_mon_t mon_complex

monitor used for complex vectors

•MHAParser::mcomplex_mon_t mon_mat_complex

monitor used for complex matrices

•MHAParser::parser_t &p_parser

parent parser to insert monitor into

Private Attributes

• bool use_mat

if true, use matrix monitor, else use vector monitor

5.12.1 Detailed Description

A class for converting AC variables to Parser monitors of correct type.

5.12.2 Constructor & Destructor Documentation

5.12.2.1 acmon::ac_monitor_t::ac_monitor_t (

MHAParser::parser_t &parent,

const std::string & name_,

algo_comm_t ac,

bool use_matrix )

Converts AC variable to parser monitor.

156 CONTENTS

Parameters

parent The parser to insert a monitor into

name_ The name of the AC variable and the monitor variable

ac Handle to algorithm communication space

use_matrix Indicates if a matrix monitor type should be used.

5.12.3 Member Function Documentation

5.12.3.1 void acmon::ac_monitor_t::getvar (

algo_comm_t ac )

Update values of monitor.

Parameters

ac Handle to algorithm communication space

5.12.4 Member Data Documentation

5.12.4.1 std::string acmon::ac_monitor_t::name

name of AC variable and parser monitor

5.12.4.2 std::string acmon::ac_monitor_t::dimstr

columns x rows

5.12.4.3 MHAParser::vﬂoat_mon_t acmon::ac_monitor_t::mon

Monitor used for real vectors.

5.12.4.4 MHAParser::mﬂoat_mon_t acmon::ac_monitor_t::mon_mat

Monitor used for real matrices.

5.12.4.5 MHAParser::vcomplex_mon_t acmon::ac_monitor_t::mon_complex

monitor used for complex vectors

5.13 acmon::acmon_t Class Reference 157

5.12.4.6 MHAParser::mcomplex_mon_t acmon::ac_monitor_t::mon_mat_complex

monitor used for complex matrices

5.12.4.7 MHAParser::parser_t& acmon::ac_monitor_t::p_parser

parent parser to insert monitor into

5.12.4.8 bool acmon::ac_monitor_t::use_mat [private]

if true, use matrix monitor, else use vector monitor

The documentation for this class was generated from the following ﬁles:

•ac_monitor_type.hh

•ac_monitor_type.cpp

5.13 acmon::acmon_t Class Reference

Inheritance diagram for acmon::acmon_t:

acmon::acmon_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

158 CONTENTS

Public Member Functions

•acmon_t (const algo_comm_t &, const std::string &, const std::string &)

•∼acmon_t ()

• void prepare (mhaconﬁg_t &)

• void release ()

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

Private Member Functions

• void save_vars ()

• void update_recmode ()

Private Attributes

•algo_comm_t ac

•MHAParser::vstring_mon_t varlist

•MHAParser::vstring_mon_t dimensions

•MHAParser::kw_t dispmode

•MHAParser::kw_t recmode

• std::vector<ac_monitor_t ∗>vars

•MHAEvents::patchbay_t<acmon_t >patchbay

• std::string chain

• std::string algo

• bool b_cont

• bool b_snapshot

Additional Inherited Members

5.13.1 Constructor & Destructor Documentation

5.13.1.1 acmon::acmon_t::acmon_t (

const algo_comm_t &iac,

const std::string & ith,

const std::string & ial )

5.13.1.2 acmon::acmon_t::∼acmon_t (

void )

5.13.2 Member Function Documentation

5.13.2.1 void acmon::acmon_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<int >(p. 730).

5.13 acmon::acmon_t Class Reference 159

5.13.2.2 void acmon::acmon_t::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.13.2.3 mha_spec_t ∗acmon::acmon_t::process (

mha_spec_t ∗s)

5.13.2.4 mha_wave_t ∗acmon::acmon_t::process (

mha_wave_t ∗s)

5.13.2.5 void acmon::acmon_t::save_vars ( ) [private]

5.13.2.6 void acmon::acmon_t::update_recmode ( ) [private]

5.13.3 Member Data Documentation

5.13.3.1 algo_comm_t acmon::acmon_t::ac [private]

5.13.3.2 MHAParser::vstring_mon_t acmon::acmon_t::varlist [private]

5.13.3.3 MHAParser::vstring_mon_t acmon::acmon_t::dimensions [private]

5.13.3.4 MHAParser::kw_t acmon::acmon_t::dispmode [private]

5.13.3.5 MHAParser::kw_t acmon::acmon_t::recmode [private]

5.13.3.6 std::vector<ac_monitor_t∗>acmon::acmon_t::vars [private]

5.13.3.7 MHAEvents::patchbay_t<acmon_t>acmon::acmon_t::patchbay [private]

5.13.3.8 std::string acmon::acmon_t::chain [private]

5.13.3.9 std::string acmon::acmon_t::algo [private]

5.13.3.10 bool acmon::acmon_t::b_cont [private]

5.13.3.11 bool acmon::acmon_t::b_snapshot [private]

The documentation for this class was generated from the following ﬁle:

•acmon.cpp

160 CONTENTS

5.14 acPooling_wave Class Reference

Inheritance diagram for acPooling_wave:

acPooling_wave

MHAPlugin::plugin_t

< acPooling_wave_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< acPooling_wave_config >

Public Member Functions

•acPooling_wave (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

Constructs our plugin.

•∼acPooling_wave ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::int_t numsamples

•MHAParser::int_t pooling_wndlen

•MHAParser::kw_t pooling_type

•MHAParser::ﬂoat_t upper_threshold

•MHAParser::ﬂoat_t lower_threshold

•MHAParser::int_t neighbourhood

5.14 acPooling_wave Class Reference 161

•MHAParser::ﬂoat_t alpha

•MHAParser::string_t p_name

•MHAParser::string_t p_biased_name

•MHAParser::string_t pool_name

•MHAParser::string_t max_pool_ind_name

•MHAParser::string_t like_ratio_name

•MHAParser::vﬂoat_t prob_bias

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<acPooling_wave >patchbay

Additional Inherited Members

5.14.1 Constructor & Destructor Documentation

5.14.1.1 acPooling_wave::acPooling_wave (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.14.1.2 acPooling_wave::∼acPooling_wave ( )

5.14.2 Member Function Documentation

5.14.2.1 mha_wave_t ∗acPooling_wave::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.14.2.2 void acPooling_wave::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

162 CONTENTS

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<acPooling_wave_conﬁg >(p. 730).

5.14.2.3 void acPooling_wave::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<acPooling_wave_conﬁg >(p. 731).

5.14.2.4 void acPooling_wave::update_cfg ( ) [private]

5.14.3 Member Data Documentation

5.14.3.1 MHAParser::int_t acPooling_wave::numsamples

5.14.3.2 MHAParser::int_t acPooling_wave::pooling_wndlen

5.14.3.3 MHAParser::kw_t acPooling_wave::pooling_type

5.14.3.4 MHAParser::ﬂoat_t acPooling_wave::upper_threshold

5.14.3.5 MHAParser::ﬂoat_t acPooling_wave::lower_threshold

5.14.3.6 MHAParser::int_t acPooling_wave::neighbourhood

5.14.3.7 MHAParser::ﬂoat_t acPooling_wave::alpha

5.14.3.8 MHAParser::string_t acPooling_wave::p_name

5.14.3.9 MHAParser::string_t acPooling_wave::p_biased_name

5.14.3.10 MHAParser::string_t acPooling_wave::pool_name

5.14.3.11 MHAParser::string_t acPooling_wave::max_pool_ind_name

5.14.3.12 MHAParser::string_t acPooling_wave::like_ratio_name

5.14.3.13 MHAParser::vﬂoat_t acPooling_wave::prob_bias

5.14.3.14 MHAEvents::patchbay_t<acPooling_wave>acPooling_wave::patchbay

[private]

The documentation for this class was generated from the following ﬁles:

•acPooling_wave.h

•acPooling_wave.cpp

5.15 acPooling_wave_conﬁg Class Reference 163

5.15 acPooling_wave_conﬁg Class Reference

Public Member Functions

•acPooling_wave_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg, acPooling←-

_wave ∗_pooling)

•∼acPooling_wave_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

• void insert ()

Public Attributes

•algo_comm_t &ac

• std::string raw_p_name

•MHA_AC::waveform_t p

•MHA_AC::waveform_t p_biased

•MHA_AC::waveform_t p_max

•MHA_AC::waveform_t like_ratio

•mha_wave_t c

• unsigned int pooling_ind

• unsigned int pooling_option

• unsigned int pooling_size

• ﬂoat up_thresh

• ﬂoat low_thresh

• int neigh

• ﬂoat alpha

•MHASignal::waveform_t pool

•MHASignal::waveform_t prob_bias_func

5.15.1 Constructor & Destructor Documentation

5.15.1.1 acPooling_wave_conﬁg::acPooling_wave_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

acPooling_wave ∗_pooling )

5.15.1.2 acPooling_wave_conﬁg::∼acPooling_wave_conﬁg ( )

5.15.2 Member Function Documentation

5.15.2.1 mha_wave_t ∗acPooling_wave_conﬁg::process (

mha_wave_t ∗wave )

5.15.2.2 void acPooling_wave_conﬁg::insert ( )

164 CONTENTS

5.15.3 Member Data Documentation

5.15.3.1 algo_comm_t& acPooling_wave_conﬁg::ac

5.15.3.2 std::string acPooling_wave_conﬁg::raw_p_name

5.15.3.3 MHA_AC::waveform_t acPooling_wave_conﬁg::p

5.15.3.4 MHA_AC::waveform_t acPooling_wave_conﬁg::p_biased

5.15.3.5 MHA_AC::waveform_t acPooling_wave_conﬁg::p_max

5.15.3.6 MHA_AC::waveform_t acPooling_wave_conﬁg::like_ratio

5.15.3.7 mha_wave_t acPooling_wave_conﬁg::c

5.15.3.8 unsigned int acPooling_wave_conﬁg::pooling_ind

5.15.3.9 unsigned int acPooling_wave_conﬁg::pooling_option

5.15.3.10 unsigned int acPooling_wave_conﬁg::pooling_size

5.15.3.11 ﬂoat acPooling_wave_conﬁg::up_thresh

5.15.3.12 ﬂoat acPooling_wave_conﬁg::low_thresh

5.15.3.13 int acPooling_wave_conﬁg::neigh

5.15.3.14 ﬂoat acPooling_wave_conﬁg::alpha

5.15.3.15 MHASignal::waveform_t acPooling_wave_conﬁg::pool

5.15.3.16 MHASignal::waveform_t acPooling_wave_conﬁg::prob_bias_func

The documentation for this class was generated from the following ﬁles:

•acPooling_wave.h

•acPooling_wave.cpp

5.16 acsave::acsave_t Class Reference 165

5.16 acsave::acsave_t Class Reference

Inheritance diagram for acsave::acsave_t:

acsave::acsave_t

MHAPlugin::plugin_t

< cfg_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< cfg_t >

Public Member Functions

•acsave_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare (mhaconﬁg_t &)

• void release ()

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void event_start_recording ()

• void event_stop_and_ﬂush ()

Private Types

• typedef std::vector<save_var_t ∗>varlist_t

Private Member Functions

• void process ()

166 CONTENTS

Private Attributes

•MHAParser::bool_t bﬂush

•MHAParser::kw_t ﬁleformat

•MHAParser::string_t fname

•MHAParser::ﬂoat_t reclen

•MHAParser::vstring_t variables

•varlist_t varlist

• std::string chain

• std::string algo

• bool b_prepared

• bool b_ﬂushed

•MHAEvents::patchbay_t<acsave_t >patchbay

Additional Inherited Members

5.16.1 Member Typedef Documentation

5.16.1.1 typedef std::vector<save_var_t∗>acsave::acsave_t::varlist_t [private]

5.16.2 Constructor & Destructor Documentation

5.16.2.1 acsave::acsave_t::acsave_t (

const algo_comm_t &iac,

const std::string & ith,

const std::string & ial )

5.16.3 Member Function Documentation

5.16.3.1 void acsave::acsave_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<cfg_t >(p. 730).

5.16.3.2 void acsave::acsave_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<cfg_t >(p. 731).

5.16 acsave::acsave_t Class Reference 167

5.16.3.3 mha_spec_t ∗acsave::acsave_t::process (

mha_spec_t ∗s)

5.16.3.4 mha_wave_t ∗acsave::acsave_t::process (

mha_wave_t ∗s)

5.16.3.5 void acsave::acsave_t::event_start_recording ( )

5.16.3.6 void acsave::acsave_t::event_stop_and_ﬂush ( )

5.16.3.7 void acsave::acsave_t::process (

void ) [private]

5.16.4 Member Data Documentation

5.16.4.1 MHAParser::bool_t acsave::acsave_t::bﬂush [private]

5.16.4.2 MHAParser::kw_t acsave::acsave_t::ﬁleformat [private]

5.16.4.3 MHAParser::string_t acsave::acsave_t::fname [private]

5.16.4.4 MHAParser::ﬂoat_t acsave::acsave_t::reclen [private]

5.16.4.5 MHAParser::vstring_t acsave::acsave_t::variables [private]

5.16.4.6 varlist_t acsave::acsave_t::varlist [private]

5.16.4.7 std::string acsave::acsave_t::chain [private]

5.16.4.8 std::string acsave::acsave_t::algo [private]

5.16.4.9 bool acsave::acsave_t::b_prepared [private]

5.16.4.10 bool acsave::acsave_t::b_ﬂushed [private]

5.16.4.11 MHAEvents::patchbay_t<acsave_t>acsave::acsave_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•acsave.cpp

168 CONTENTS

5.17 acsave::cfg_t Class Reference

Public Member Functions

•cfg_t (const algo_comm_t &iac, unsigned int imax_frames, std::vector<std::string >

&var_names)

•∼cfg_t ()

• void store_frame ()

• void ﬂush_data (const std::string &, unsigned int)

Private Attributes

•algo_comm_t ac

• unsigned int nvars

•save_var_t ∗∗ varlist

• unsigned int rec_frames

• unsigned int max_frames

5.17.1 Constructor & Destructor Documentation

5.17.1.1 cfg_t::cfg_t (

const algo_comm_t &iac,

unsigned int imax_frames,

std::vector<std::string >&var_names )

5.17.1.2 cfg_t::∼cfg_t ( )

5.17.2 Member Function Documentation

5.17.2.1 void cfg_t::store_frame ( )

This function is called in the processing thread.

5.17.2.2 void cfg_t::ﬂush_data (

const std::string & ﬁlename,

unsigned int fmt )

This function is called in the conﬁguration thread.

Parameters

ﬁlename Output ﬁle name

fmt Output ﬁle format

5.18 acsave::mat4head_t Struct Reference 169

5.17.3 Member Data Documentation

5.17.3.1 algo_comm_t acsave::cfg_t::ac [private]

5.17.3.2 unsigned int acsave::cfg_t::nvars [private]

5.17.3.3 save_var_t∗∗ acsave::cfg_t::varlist [private]

5.17.3.4 unsigned int acsave::cfg_t::rec_frames [private]

5.17.3.5 unsigned int acsave::cfg_t::max_frames [private]

The documentation for this class was generated from the following ﬁle:

•acsave.cpp

5.18 acsave::mat4head_t Struct Reference

Public Attributes

• int32_t t

• int32_t rows

• int32_t cols

• int32_t imag

• int32_t namelen

5.18.1 Member Data Documentation

5.18.1.1 int32_t acsave::mat4head_t::t

5.18.1.2 int32_t acsave::mat4head_t::rows

5.18.1.3 int32_t acsave::mat4head_t::cols

5.18.1.4 int32_t acsave::mat4head_t::imag

5.18.1.5 int32_t acsave::mat4head_t::namelen

The documentation for this struct was generated from the following ﬁle:

•acsave.cpp

170 CONTENTS

5.19 acsave::save_var_t Class Reference

Public Member Functions

•save_var_t (const std::string &, int, const algo_comm_t &)

•∼save_var_t ()

• void store_frame ()

• void save_txt (FILE ∗, unsigned int)

• void save_mat4 (FILE ∗, unsigned int)

• void save_m (FILE ∗, unsigned int)

Public Attributes

• double ∗data

Private Attributes

• std::string name

• unsigned int nframes

• unsigned int ndim

• unsigned int maxframe

•algo_comm_t ac

• unsigned int framecnt

• bool b_complex

5.19.1 Constructor & Destructor Documentation

5.19.1.1 acsave::save_var_t::save_var_t (

const std::string & nm,

int n,

const algo_comm_t &iac )

5.19.1.2 acsave::save_var_t::∼save_var_t ( )

5.19.2 Member Function Documentation

5.19.2.1 void acsave::save_var_t::store_frame ( )

5.19.2.2 void acsave::save_var_t::save_txt (

FILE ∗fh,

unsigned int writeframes )

5.19 acsave::save_var_t Class Reference 171

5.19.2.3 void acsave::save_var_t::save_mat4 (

FILE ∗fh,

unsigned int writeframes )

5.19.2.4 void acsave::save_var_t::save_m (

FILE ∗fh,

unsigned int writeframes )

5.19.3 Member Data Documentation

5.19.3.1 double∗acsave::save_var_t::data

5.19.3.2 std::string acsave::save_var_t::name [private]

5.19.3.3 unsigned int acsave::save_var_t::nframes [private]

5.19.3.4 unsigned int acsave::save_var_t::ndim [private]

5.19.3.5 unsigned int acsave::save_var_t::maxframe [private]

5.19.3.6 algo_comm_t acsave::save_var_t::ac [private]

5.19.3.7 unsigned int acsave::save_var_t::framecnt [private]

5.19.3.8 bool acsave::save_var_t::b_complex [private]

The documentation for this class was generated from the following ﬁle:

•acsave.cpp

172 CONTENTS

5.20 acSteer Class Reference

Inheritance diagram for acSteer:

acSteer

MHAPlugin::plugin_t

< acSteer_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< acSteer_config >

Public Member Functions

•acSteer (algo_comm_t &ac, const std::string &chain_name, const std::string &algo_←-

name)

Constructs our plugin.

•∼acSteer ()

•mha_spec_t ∗process (mha_spec_t ∗)

Thos method is a NOOP.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::string_t steerFile

•MHAParser::string_t acSteerName1

•MHAParser::string_t acSteerName2

•MHAParser::int_t nsteerchan

•MHAParser::int_t nrefmic

5.20 acSteer Class Reference 173

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<acSteer >patchbay

Additional Inherited Members

5.20.1 Constructor & Destructor Documentation

5.20.1.1 acSteer::acSteer (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.20.1.2 acSteer::∼acSteer ( )

5.20.2 Member Function Documentation

5.20.2.1 mha_spec_t ∗acSteer::process (

mha_spec_t ∗signal )

Thos method is a NOOP.

5.20.2.2 void acSteer::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<acSteer_conﬁg >(p. 730).

174 CONTENTS

5.20.2.3 void acSteer::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<acSteer_conﬁg >(p. 731).

5.20.2.4 void acSteer::update_cfg ( ) [private]

5.20.3 Member Data Documentation

5.20.3.1 MHAParser::string_t acSteer::steerFile

5.20.3.2 MHAParser::string_t acSteer::acSteerName1

5.20.3.3 MHAParser::string_t acSteer::acSteerName2

5.20.3.4 MHAParser::int_t acSteer::nsteerchan

5.20.3.5 MHAParser::int_t acSteer::nrefmic

5.20.3.6 MHAEvents::patchbay_t<acSteer>acSteer::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•acSteer.h

•acSteer.cpp

5.21 acSteer_conﬁg Class Reference

Public Member Functions

•acSteer_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg, acSteer ∗acSteer)

•∼acSteer_conﬁg ()

• void insert ()

Public Attributes

• unsigned int nchan

• unsigned int nfreq

• unsigned int nsteerchan

• unsigned int nrefmic

• unsigned int nangle

•MHA_AC::spectrum_t specSteer1

•MHA_AC::spectrum_t specSteer2

5.21 acSteer_conﬁg Class Reference 175

5.21.1 Constructor & Destructor Documentation

5.21.1.1 acSteer_conﬁg::acSteer_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

acSteer ∗acSteer )

5.21.1.2 acSteer_conﬁg::∼acSteer_conﬁg ( )

5.21.2 Member Function Documentation

5.21.2.1 void acSteer_conﬁg::insert ( )

5.21.3 Member Data Documentation

5.21.3.1 unsigned int acSteer_conﬁg::nchan

5.21.3.2 unsigned int acSteer_conﬁg::nfreq

5.21.3.3 unsigned int acSteer_conﬁg::nsteerchan

5.21.3.4 unsigned int acSteer_conﬁg::nrefmic

5.21.3.5 unsigned int acSteer_conﬁg::nangle

5.21.3.6 MHA_AC::spectrum_t acSteer_conﬁg::specSteer1

5.21.3.7 MHA_AC::spectrum_t acSteer_conﬁg::specSteer2

The documentation for this class was generated from the following ﬁles:

•acSteer.h

•acSteer.cpp

176 CONTENTS

5.22 acTransform_wave Class Reference

Inheritance diagram for acTransform_wave:

acTransform_wave

MHAPlugin::plugin_t

< acTransform_wave_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< acTransform_wave_config >

Public Member Functions

•acTransform_wave (algo_comm_t &ac, const std::string &chain_name, const std←-

::string &algo_name)

Constructs our plugin.

•∼acTransform_wave ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::string_t ang_name

•MHAParser::string_t raw_p_name

•MHAParser::string_t raw_p_max_name

•MHAParser::string_t rotated_p_name

•MHAParser::string_t rotated_p_max_name

•MHAParser::int_t numsamples

•MHAParser::bool_t to_from

5.22 acTransform_wave Class Reference 177

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<acTransform_wave >patchbay

Additional Inherited Members

5.22.1 Constructor & Destructor Documentation

5.22.1.1 acTransform_wave::acTransform_wave (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.22.1.2 acTransform_wave::∼acTransform_wave ( )

5.22.2 Member Function Documentation

5.22.2.1 mha_wave_t ∗acTransform_wave::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.22.2.2 void acTransform_wave::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<acTransform_wave_conﬁg >(p. 730).

178 CONTENTS

5.22.2.3 void acTransform_wave::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<acTransform_wave_conﬁg >(p. 731).

5.22.2.4 void acTransform_wave::update_cfg ( ) [private]

5.22.3 Member Data Documentation

5.22.3.1 MHAParser::string_t acTransform_wave::ang_name

5.22.3.2 MHAParser::string_t acTransform_wave::raw_p_name

5.22.3.3 MHAParser::string_t acTransform_wave::raw_p_max_name

5.22.3.4 MHAParser::string_t acTransform_wave::rotated_p_name

5.22.3.5 MHAParser::string_t acTransform_wave::rotated_p_max_name

5.22.3.6 MHAParser::int_t acTransform_wave::numsamples

5.22.3.7 MHAParser::bool_t acTransform_wave::to_from

5.22.3.8 MHAEvents::patchbay_t<acTransform_wave>acTransform_wave::patchbay

[private]

The documentation for this class was generated from the following ﬁles:

•acTransform_wave.h

•acTransform_wave.cpp

5.23 acTransform_wave_conﬁg Class Reference

Public Member Functions

•acTransform_wave_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg, ac←-

Transform_wave ∗_transform)

•∼acTransform_wave_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

5.23 acTransform_wave_conﬁg Class Reference 179

Public Attributes

•algo_comm_t &ac

• std::string ang_name

• std::string raw_p_name

• std::string raw_p_max_name

•MHA_AC::waveform_t rotated_p

•MHA_AC::int_t rotated_i

• unsigned int offset

• unsigned int resolution

• unsigned int to_from

5.23.1 Constructor & Destructor Documentation

5.23.1.1 acTransform_wave_conﬁg::acTransform_wave_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

acTransform_wave ∗_transform )

5.23.1.2 acTransform_wave_conﬁg::∼acTransform_wave_conﬁg ( )

5.23.2 Member Function Documentation

5.23.2.1 mha_wave_t ∗acTransform_wave_conﬁg::process (

mha_wave_t ∗wave )

5.23.3 Member Data Documentation

5.23.3.1 algo_comm_t& acTransform_wave_conﬁg::ac

5.23.3.2 std::string acTransform_wave_conﬁg::ang_name

5.23.3.3 std::string acTransform_wave_conﬁg::raw_p_name

5.23.3.4 std::string acTransform_wave_conﬁg::raw_p_max_name

5.23.3.5 MHA_AC::waveform_t acTransform_wave_conﬁg::rotated_p

5.23.3.6 MHA_AC::int_t acTransform_wave_conﬁg::rotated_i

5.23.3.7 unsigned int acTransform_wave_conﬁg::offset

5.23.3.8 unsigned int acTransform_wave_conﬁg::resolution

5.23.3.9 unsigned int acTransform_wave_conﬁg::to_from

The documentation for this class was generated from the following ﬁles:

•acTransform_wave.h

•acTransform_wave.cpp

180 CONTENTS

5.24 addsndﬁle::addsndﬁle_if_t Class Reference

Inheritance diagram for addsndﬁle::addsndﬁle_if_t:

addsndfile::addsndfile_if_t

wave_reader

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< runtime_cfg_t >

level_adaptor

Public Member Functions

•addsndﬁle_if_t (algo_comm_t, const char ∗, const char ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void update ()

• void change_mode ()

• void set_level ()

• void scan_dir ()

Private Attributes

•MHAParser::string_t ﬁlename

•MHAParser::string_t path

•MHAParser::bool_t loop

•MHAParser::ﬂoat_t level

5.24 addsndﬁle::addsndﬁle_if_t Class Reference 181

•MHAParser::kw_t levelmode

•MHAParser::kw_t resamplingmode

•MHAParser::vint_t channels

•MHAParser::kw_t mode

•MHAParser::ﬂoat_t ramplen

•MHAParser::int_t startpos

•MHAParser::vint_mon_t mapping

•MHAParser::int_mon_t numchannels

•MHAParser::int_mon_t mhachannels

•MHAParser::int_mon_t active

•MHAParser::string_t search_pattern

•MHAParser::vstring_mon_t search_result

• unsigned int uint_mode

•MHAEvents::patchbay_t<addsndﬁle_if_t >patchbay

Additional Inherited Members

5.24.1 Constructor & Destructor Documentation

5.24.1.1 addsndﬁle::addsndﬁle_if_t::addsndﬁle_if_t (

algo_comm_t iac,

const char ∗,

const char ∗)

5.24.2 Member Function Documentation

5.24.2.1 mha_wave_t ∗addsndﬁle::addsndﬁle_if_t::process (

mha_wave_t ∗s)

5.24.2.2 void addsndﬁle::addsndﬁle_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<runtime_cfg_t >(p. 730).

5.24.2.3 void addsndﬁle::addsndﬁle_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<runtime_cfg_t >(p. 731).

182 CONTENTS

5.24.2.4 void addsndﬁle::addsndﬁle_if_t::update ( ) [private]

5.24.2.5 void addsndﬁle::addsndﬁle_if_t::change_mode ( ) [private]

5.24.2.6 void addsndﬁle::addsndﬁle_if_t::set_level ( ) [private]

5.24.2.7 void addsndﬁle::addsndﬁle_if_t::scan_dir ( ) [private]

5.24.3 Member Data Documentation

5.24.3.1 MHAParser::string_t addsndﬁle::addsndﬁle_if_t::ﬁlename [private]

5.24.3.2 MHAParser::string_t addsndﬁle::addsndﬁle_if_t::path [private]

5.24.3.3 MHAParser::bool_t addsndﬁle::addsndﬁle_if_t::loop [private]

5.24.3.4 MHAParser::ﬂoat_t addsndﬁle::addsndﬁle_if_t::level [private]

5.24.3.5 MHAParser::kw_t addsndﬁle::addsndﬁle_if_t::levelmode [private]

5.24.3.6 MHAParser::kw_t addsndﬁle::addsndﬁle_if_t::resamplingmode [private]

5.24.3.7 MHAParser::vint_t addsndﬁle::addsndﬁle_if_t::channels [private]

5.24.3.8 MHAParser::kw_t addsndﬁle::addsndﬁle_if_t::mode [private]

5.24.3.9 MHAParser::ﬂoat_t addsndﬁle::addsndﬁle_if_t::ramplen [private]

5.24.3.10 MHAParser::int_t addsndﬁle::addsndﬁle_if_t::startpos [private]

5.24.3.11 MHAParser::vint_mon_t addsndﬁle::addsndﬁle_if_t::mapping [private]

5.24.3.12 MHAParser::int_mon_t addsndﬁle::addsndﬁle_if_t::numchannels [private]

5.24.3.13 MHAParser::int_mon_t addsndﬁle::addsndﬁle_if_t::mhachannels [private]

5.24.3.14 MHAParser::int_mon_t addsndﬁle::addsndﬁle_if_t::active [private]

5.24.3.15 MHAParser::string_t addsndﬁle::addsndﬁle_if_t::search_pattern [private]

5.24.3.16 MHAParser::vstring_mon_t addsndﬁle::addsndﬁle_if_t::search_result [private]

5.24.3.17 unsigned int addsndﬁle::addsndﬁle_if_t::uint_mode [private]

5.24.3.18 MHAEvents::patchbay_t<addsndﬁle_if_t>addsndﬁle::addsndﬁle_if_t::patchbay

[private]

The documentation for this class was generated from the following ﬁle:

•addsndﬁle.cpp

5.25 addsndﬁle::level_adapt_t Class Reference 183

5.25 addsndﬁle::level_adapt_t Class Reference

Inheritance diagram for addsndﬁle::level_adapt_t:

addsndfile::level_adapt_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•level_adapt_t (mhaconﬁg_t cf, mha_real_t adapt_len, mha_real_t l_new_, mha_real←-

_t l_old_)

• void update_frame ()

•mha_real_t get_level () const

• bool can_update () const

Private Attributes

• unsigned int ilen

• unsigned int pos

•MHAWindow::fun_t wnd

•mha_real_t l_new

•mha_real_t l_old

Additional Inherited Members

5.25.1 Constructor & Destructor Documentation

5.25.1.1 addsndﬁle::level_adapt_t::level_adapt_t (

mhaconﬁg_t cf,

mha_real_t adapt_len,

mha_real_t l_new_,

mha_real_t l_old_ )

184 CONTENTS

5.25.2 Member Function Documentation

5.25.2.1 void addsndﬁle::level_adapt_t::update_frame ( )

5.25.2.2 mha_real_t addsndﬁle::level_adapt_t::get_level ( ) const [inline]

5.25.2.3 bool addsndﬁle::level_adapt_t::can_update ( ) const [inline]

5.25.3 Member Data Documentation

5.25.3.1 unsigned int addsndﬁle::level_adapt_t::ilen [private]

5.25.3.2 unsigned int addsndﬁle::level_adapt_t::pos [private]

5.25.3.3 MHAWindow::fun_t addsndﬁle::level_adapt_t::wnd [private]

5.25.3.4 mha_real_t addsndﬁle::level_adapt_t::l_new [private]

5.25.3.5 mha_real_t addsndﬁle::level_adapt_t::l_old [private]

The documentation for this class was generated from the following ﬁle:

•addsndﬁle.cpp

5.26 addsndﬁle::resampled_soundﬁle_t Class Reference

Reads sound from ﬁle and resamples it if necessary and wanted.

Inheritance diagram for addsndﬁle::resampled_soundﬁle_t:

addsndfile::resampled

_soundfile_t

MHASndFile::sf_wave_t

MHASndFile::sf_t MHASignal::waveform_t

addsndfile::waveform

_proxy_t

mha_wave_t

5.26 addsndﬁle::resampled_soundﬁle_t Class Reference 185

Public Member Functions

•resampled_soundﬁle_t (const std::string &name, ﬂoat mha_sampling_rate, addsndﬁle←-

_resampling_mode_t resampling_mode)

Reads sound from ﬁle and resamples if necessary and wanted.

Additional Inherited Members

5.26.1 Detailed Description

Reads sound from ﬁle and resamples it if necessary and wanted.

Sound data can then be used by addsndﬁle.

5.26.2 Constructor & Destructor Documentation

5.26.2.1 addsndﬁle::resampled_soundﬁle_t::resampled_soundﬁle_t (

const std::string & name,

ﬂoat mha_sampling_rate,

addsndﬁle_resampling_mode_t resampling_mode )

Reads sound from ﬁle and resamples if necessary and wanted.

If the sound ﬁle does not specify a sampling rate, then the sound data is always used without

resampling.

Parameters

name Sound ﬁle name

mha_sampling_rate The sampling rate of the MHA signal processing at the point of the

addsndﬁle plugin

resampling_mode DONT_RESAMPLE_STRICT: Do not resample, just use the samples

from the sound ﬁle at the current sample rate, even if the sample rate

of the sound ﬁle differs. DONT_RESAMPLE_PERMISSIVE: Do not

resample, if the sample rate of the MHA differs from the sample rate

of the sound ﬁle, raise an error. DO_RESAMPLE: Resample.

Exceptions

MHA_Error (p. 445)If the sampling rate of the ﬁle does not match the sampling rate of

the MHA and DONT_RESAMPLE_STRICT was requested. If

resampling failed (e.g. due to non-rational quotient of MHA sampling

rate and sound ﬁle sampling rate).

186 CONTENTS

The documentation for this class was generated from the following ﬁle:

•addsndﬁle.cpp

5.27 addsndﬁle::sndﬁle_t Class Reference

Inheritance diagram for addsndﬁle::sndﬁle_t:

addsndfile::sndfile_t

MHASignal::loop_wavefragment_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•sndﬁle_t (const std::string &name, bool loop, unsigned int level_mode, std::vector<

int >channels_, unsigned int nchannels, std::vector<int >&mapping, int &numchan-

nels, unsigned int startpos, ﬂoat mha_sampling_rate, addsndﬁle_resampling_mode_t

resampling_mode)

Additional Inherited Members

5.27.1 Constructor & Destructor Documentation

5.28 addsndﬁle::waveform_proxy_t Class Reference 187

5.27.1.1 addsndﬁle::sndﬁle_t::sndﬁle_t (

const std::string & name,

bool loop,

unsigned int level_mode,

std::vector<int >channels_,

unsigned int nchannels,

std::vector<int >&mapping,

int & numchannels,

unsigned int startpos,

ﬂoat mha_sampling_rate,

addsndﬁle_resampling_mode_t resampling_mode )

The documentation for this class was generated from the following ﬁle:

•addsndﬁle.cpp

5.28 addsndﬁle::waveform_proxy_t Class Reference

Class helps to specify which instance of MHASignal_waveform_t parent instance is meant in

resampled_soundﬁle_t (p. 184).

Inheritance diagram for addsndﬁle::waveform_proxy_t:

addsndfile::waveform

_proxy_t

addsndfile::resampled

_soundfile_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•waveform_proxy_t (unsigned frames, unsigned channels)

188 CONTENTS

Additional Inherited Members

5.28.1 Detailed Description

Class helps to specify which instance of MHASignal_waveform_t parent instance is meant in

resampled_soundﬁle_t (p. 184).

5.28.2 Constructor & Destructor Documentation

5.28.2.1 addsndﬁle::waveform_proxy_t::waveform_proxy_t (

unsigned frames,

unsigned channels )[inline]

The documentation for this class was generated from the following ﬁle:

•addsndﬁle.cpp

5.29 ADM::ADM<F>Class Template Reference

Adaptive differential microphone, working for speech frequency range.

Public Member Functions

•ADM (F fs, F dist, unsigned lp_order, const F ∗lp_alphas, unsigned decomb_order, const

F∗decomb_alphas, F tau_beta=F(50e-3), F mu_beta=F(1e-4))

Create Adaptive Differential Microphone.

• F process (const F &front, const F &back, const F &external_beta=F(-1))

ADM (p. 188) processes one frame.

• F beta () const

Private Attributes

•Delay<F>m_delay_front

•Delay<F>m_delay_back

•Linearphase_FIR<F>m_lp_bf

•Linearphase_FIR<F>m_lp_result

•Linearphase_FIR<F>m_decomb

• F m_beta

• F m_mu_beta

• F m_powerﬁlter_coeff

• F m_powerﬁlter_norm

• F m_powerﬁlter_state

5.29 ADM::ADM<F>Class Template Reference 189

5.29.1 Detailed Description

template<class F>

class ADM::ADM<F>

Adaptive differential microphone, working for speech frequency range.

5.29.2 Constructor & Destructor Documentation

5.29.2.1 template<class F >ADM::ADM<F>::ADM (

Ffs,

Fdist,

unsigned lp_order,

const F ∗lp_alphas,

unsigned decomb_order,

const F ∗decomb_alphas,

Ftau_beta = F(50e-3),

Fmu_beta = F(1e-4) )

Create Adaptive Differential Microphone.

Parameters

fs Sampling rate / Hz

dist Distance between physical microphones / m

lp_order Filter order of FIR lowpass ﬁlter used for adaptation

lp_alphas Pointer to array of alpha coefﬁcients for the lowpass ﬁlter used for

adaptation. Since this class uses linear phase FIR ﬁlters only, only the

ﬁrst half (order/2 + 1) of the coefﬁcients will be read (coefﬁcients for

linear-phase FIR ﬁlters are symmetric).

decomb_order Filter order of FIR compensation ﬁlter (compensates for comb ﬁlter

characteristic)

decomb_alphas Pointer to array of alpha coefﬁcients for the compensation ﬁlter used to

compensate for the comb ﬁlter characteristic. Since this class uses linear

phase FIR ﬁlters only, only the ﬁrst half (order/2 + 1)of the coefﬁcients

will be read (coefﬁcients for linear-phase FIR ﬁlters are symmetric).

tau_beta Time constant of the lowpass ﬁlter used for averaging the power of the

output signal

mu_beta adaption speed

5.29.3 Member Function Documentation

190 CONTENTS

5.29.3.1 template<class F >FADM::ADM<F>::process (

const F & front,

const F & back,

const F & external_beta = F(-1) )[inline]

ADM (p. 188) processes one frame.

Parameters

front The current front input signal sample

back The current rear input signal sample

external_beta If >= 0, this is used as the "beta" parameter for direction to ﬁlter out. Else,

the beta parameter is adapted to ﬁltered out a direction so that best

reduction of signal intensity from the back hemisphere is achieved.

Returns

The computed output sample

5.29.3.2 template<class F >FADM::ADM<F>::beta ( ) const [inline]

5.29.4 Member Data Documentation

5.29.4.1 template<class F >Delay<F>ADM::ADM<F>::m_delay_front [private]

5.29.4.2 template<class F >Delay<F>ADM::ADM<F>::m_delay_back [private]

5.29.4.3 template<class F >Linearphase_FIR<F>ADM::ADM<F>::m_lp_bf [private]

5.29.4.4 template<class F >Linearphase_FIR<F>ADM::ADM<F>::m_lp_result [private]

5.29.4.5 template<class F >Linearphase_FIR<F>ADM::ADM<F>::m_decomb [private]

5.29.4.6 template<class F >FADM::ADM<F>::m_beta [private]

5.29.4.7 template<class F >FADM::ADM<F>::m_mu_beta [private]

5.29.4.8 template<class F >FADM::ADM<F>::m_powerﬁlter_coeff [private]

5.29.4.9 template<class F >FADM::ADM<F>::m_powerﬁlter_norm [private]

5.29.4.10 template<class F >FADM::ADM<F>::m_powerﬁlter_state [private]

The documentation for this class was generated from the following ﬁle:

•adm.hh

5.30 ADM::Delay<F>Class Template Reference 191

5.30 ADM::Delay<F>Class Template Reference

A delay-line class which can also do subsample-delays for a limited frequency range below

fs/4.

Public Member Functions

•Delay (F samples, F f_design, F fs)

Create a signal delay object.

•∼Delay ()

• F process (const F &in_sample)

Apply delay to signal.

Private Attributes

• unsigned m_fullsamples

Integer part of delay.

• F m_coeff

coefﬁcient for 1st order IIR lowpass ﬁlter which does the subsample delay

• F m_norm

normalization for the IIR subsample delay ﬁlter

• F ∗m_state

Ringbuffer: Delayline.

• unsigned m_now_in

current position for inserting new samples into m_state ringbuffer

5.30.1 Detailed Description

template<class F>

class ADM::Delay<F>

A delay-line class which can also do subsample-delays for a limited frequency range below

fs/4.

5.30.2 Constructor & Destructor Documentation

5.30.2.1 template<class F >ADM::Delay<F>::Delay (

Fsamples,

Ff_design,

Ffs )

Create a signal delay object.

192 CONTENTS

Parameters

samples number of samples to delay (may be non-integer)

f_design subsampledelay is exact for this frequency

fs sampling frequency

5.30.2.2 template<class F >ADM::Delay<F>::∼Delay ( )

5.30.3 Member Function Documentation

5.30.3.1 template<class F >FADM::Delay<F>::process (

const F & in_sample )[inline]

Apply delay to signal.

Parameters

in_sample The current input signal sample

Returns

The computed output sample

5.30.4 Member Data Documentation

5.30.4.1 template<class F >unsigned ADM::Delay<F>::m_fullsamples [private]

Integer part of delay.

5.30.4.2 template<class F >FADM::Delay<F>::m_coeff [private]

coefﬁcient for 1st order IIR lowpass ﬁlter which does the subsample delay

5.30.4.3 template<class F >FADM::Delay<F>::m_norm [private]

normalization for the IIR subsample delay ﬁlter

5.30.4.4 template<class F >F∗ADM::Delay<F>::m_state [private]

Ringbuffer: Delayline.

5.31 ADM::Linearphase_FIR<F>Class Template Reference 193

5.30.4.5 template<class F >unsigned ADM::Delay<F>::m_now_in [private]

current position for inserting new samples into m_state ringbuffer

The documentation for this class was generated from the following ﬁle:

•adm.hh

5.31 ADM::Linearphase_FIR<F>Class Template Reference

An efﬁcient linear-phase ﬁr ﬁlter implementation.

Public Member Functions

•Linearphase_FIR (unsigned order, const F ∗alphas)

Create linear-phase FIR ﬁlter.

•∼Linearphase_FIR ()

• F process (const F &in_sample)

Filter one sample with this linear-phase FIR ﬁlter.

Private Attributes

• unsigned m_order

The ﬁlter order of this linear-phase FIR ﬁlter.

• F ∗m_alphas

FIR ﬁlter coefﬁcients.

• F ∗m_output

Ringbuffer for building future output.

• unsigned m_now

current start of ringbuffer

5.31.1 Detailed Description

template<class F>

class ADM::Linearphase_FIR<F>

An efﬁcient linear-phase ﬁr ﬁlter implementation.

5.31.2 Constructor & Destructor Documentation

5.31.2.1 template<class F >ADM::Linearphase_FIR<F>::Linearphase_FIR (

unsigned order,

const F ∗alphas )

Create linear-phase FIR ﬁlter.

194 CONTENTS

Parameters

order ﬁlter order of this FIR ﬁlter. restriction: must be even.

alphas pointer to Array of alpha coefﬁcients. Since this class is for linear phase FIR ﬁlters

only, only (order / 2 + 1) coefﬁcients will be read. (Coefﬁcients for linear-phase FIR

ﬁlters are symmetric.)

5.31.2.2 template<class F >ADM::Linearphase_FIR<F>::∼Linearphase_FIR ( )

5.31.3 Member Function Documentation

5.31.3.1 template<class F >FADM::Linearphase_FIR<F>::process (

const F & in_sample )[inline]

Filter one sample with this linear-phase FIR ﬁlter.

Parameters

in_sample the current input sample

Returns

the computed output sample

5.31.4 Member Data Documentation

5.31.4.1 template<class F >unsigned ADM::Linearphase_FIR<F>::m_order [private]

The ﬁlter order of this linear-phase FIR ﬁlter.

5.31.4.2 template<class F >F∗ADM::Linearphase_FIR<F>::m_alphas [private]

FIR ﬁlter coefﬁcients.

Only m_order / 2 + 1 coefﬁcients need to be stored since coefﬁcients of linear-phase FIR ﬁlters

are symmetric

5.31.4.3 template<class F >F∗ADM::Linearphase_FIR<F>::m_output [private]

Ringbuffer for building future output.

5.32 adm_if_t Class Reference 195

5.31.4.4 template<class F >unsigned ADM::Linearphase_FIR<F>::m_now [private]

current start of ringbuffer

The documentation for this class was generated from the following ﬁle:

•adm.hh

5.32 adm_if_t Class Reference

Inheritance diagram for adm_if_t:

adm_if_t

MHAPlugin::plugin_t

< adm_rtconfig_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< adm_rtconfig_t >

Public Member Functions

•adm_if_t (const algo_comm_t &ac, const std::string &thread_name, const std::string

&algo_name)

•mha_wave_t ∗process (mha_wave_t ∗in)

• virtual void prepare (mhaconﬁg_t &)

• virtual void release ()

Private Member Functions

• void update ()

• bool is_prepared ()

196 CONTENTS

Private Attributes

•MHASignal::waveform_t ∗out

•MHAParser::vint_t front_channels

•MHAParser::vint_t rear_channels

•MHAParser::vﬂoat_t distances

•MHAParser::int_t lp_order

•MHAParser::int_t decomb_order

•MHAParser::int_t bypass

•MHAParser::ﬂoat_t beta

•MHAParser::vﬂoat_t mu_beta

•MHAParser::vﬂoat_t tau_beta

•MHAParser::vﬂoat_mon_t coeff_lp

•MHAParser::vﬂoat_mon_t coeff_decomb

• unsigned input_channels

•mha_real_t srate

•MHAEvents::patchbay_t<adm_if_t >patchbay

Additional Inherited Members

5.32.1 Constructor & Destructor Documentation

5.32.1.1 adm_if_t::adm_if_t (

const algo_comm_t &ac,

const std::string & thread_name,

const std::string & algo_name )

5.32.2 Member Function Documentation

5.32.2.1 mha_wave_t ∗adm_if_t::process (

mha_wave_t ∗in )

5.32.2.2 void adm_if_t::prepare (

mhaconﬁg_t &cfg )[virtual]

Implements MHAPlugin::plugin_t<adm_rtconﬁg_t >(p. 730).

5.32.2.3 void adm_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<adm_rtconﬁg_t >(p. 731).

5.33 adm_rtconﬁg_t Class Reference 197

5.32.2.4 void adm_if_t::update ( ) [private]

5.32.2.5 bool adm_if_t::is_prepared ( ) [inline],[private]

5.32.3 Member Data Documentation

5.32.3.1 MHASignal::waveform_t∗adm_if_t::out [private]

5.32.3.2 MHAParser::vint_t adm_if_t::front_channels [private]

5.32.3.3 MHAParser::vint_t adm_if_t::rear_channels [private]

5.32.3.4 MHAParser::vﬂoat_t adm_if_t::distances [private]

5.32.3.5 MHAParser::int_t adm_if_t::lp_order [private]

5.32.3.6 MHAParser::int_t adm_if_t::decomb_order [private]

5.32.3.7 MHAParser::int_t adm_if_t::bypass [private]

5.32.3.8 MHAParser::ﬂoat_t adm_if_t::beta [private]

5.32.3.9 MHAParser::vﬂoat_t adm_if_t::mu_beta [private]

5.32.3.10 MHAParser::vﬂoat_t adm_if_t::tau_beta [private]

5.32.3.11 MHAParser::vﬂoat_mon_t adm_if_t::coeff_lp [private]

5.32.3.12 MHAParser::vﬂoat_mon_t adm_if_t::coeff_decomb [private]

5.32.3.13 unsigned adm_if_t::input_channels [private]

5.32.3.14 mha_real_t adm_if_t::srate [private]

5.32.3.15 MHAEvents::patchbay_t<adm_if_t>adm_if_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•adm.cpp

5.33 adm_rtconﬁg_t Class Reference

Public Types

• typedef ADM::ADM<mha_real_t >adm_t

198 CONTENTS

Public Member Functions

•adm_rtconﬁg_t (unsigned nchannels_in, unsigned nchannels_out, const std::vector<

int >&front_channels, const std::vector<int >&rear_channels, const mha_real_t fs,

const std::vector<mha_real_t >&distances, const int lp_order, const int decomb_order,

const std::vector<mha_real_t >&tau_beta, const std::vector<mha_real_t >&mu_←-

beta)

Construct new ADMs.

• virtual ∼adm_rtconﬁg_t ()

• size_t num_adms () const

•adm_t &adm (unsigned index)

Returns adm object number index.

• int front_channel (unsigned index) const

Returns index of front channel for adm number index.

• int rear_channel (unsigned index) const

Returns index of rear channel for adm number index.

Private Member Functions

• void check_index (unsigned index) const

Index checking for all internal arrays.

Private Attributes

• std::vector<int >front_channels

Indices of channels containing the signals from the front microphones.

• std::vector<int >rear_channels

Indices of channels containing the signals from the rear microphones.

•MHASignal::waveform_t ∗lp_coeffs

Lowpass ﬁlter coefﬁcients.

• std::vector<MHASignal::waveform_t ∗>decomb_coeffs

Decomb-Filter coefﬁcients.

• std::vector<adm_t ∗>adms

ADMs.

5.33.1 Member Typedef Documentation

5.33.1.1 typedef ADM::ADM<mha_real_t>adm_rtconﬁg_t::adm_t

5.33.2 Constructor & Destructor Documentation

5.33 adm_rtconﬁg_t Class Reference 199

5.33.2.1 adm_rtconﬁg_t::adm_rtconﬁg_t (

unsigned nchannels_in,

unsigned nchannels_out,

const std::vector<int >&front_channels,

const std::vector<int >&rear_channels,

const mha_real_t fs,

const std::vector<mha_real_t >&distances,

const int lp_order,

const int decomb_order,

const std::vector<mha_real_t >&tau_beta,

const std::vector<mha_real_t >&mu_beta )

Construct new ADMs.

Used when conﬁguration changes.

Parameters

nchannels_in Number of input channels

nchannels_out Number of output channels

front_channels Parser's front_channels setting

rear_channels Parser's front_channels setting

fs Sampling rate / Hz

distances Distances between microphones / m

lp_order Filter order of FIR lowpass ﬁlter for adaptation

decomb_order Filter order of FIR compensation ﬁlter (compensates for comb ﬁlter

characteristic)

tau_beta Time constants of the lowpass ﬁlter used for averaging the power of the

output signal used for adaptation

mu_beta Adaptation step sizes

5.33.2.2 adm_rtconﬁg_t::∼adm_rtconﬁg_t ( ) [virtual]

5.33.3 Member Function Documentation

5.33.3.1 void adm_rtconﬁg_t::check_index (

unsigned index ) const [inline],[private]

Index checking for all internal arrays.

Exceptions

MHA_Error (p. 445)if index out of range.

200 CONTENTS

5.33.3.2 size_t adm_rtconﬁg_t::num_adms ( ) const [inline]

5.33.3.3 adm_t& adm_rtconﬁg_t::adm (

unsigned index )[inline]

Returns adm object number index.

5.33.3.4 int adm_rtconﬁg_t::front_channel (

unsigned index ) const [inline]

Returns index of front channel for adm number index.

5.33.3.5 int adm_rtconﬁg_t::rear_channel (

unsigned index ) const [inline]

Returns index of rear channel for adm number index.

5.33.4 Member Data Documentation

5.33.4.1 std::vector<int>adm_rtconﬁg_t::front_channels [private]

Indices of channels containing the signals from the front microphones.

5.33.4.2 std::vector<int>adm_rtconﬁg_t::rear_channels [private]

Indices of channels containing the signals from the rear microphones.

5.33.4.3 MHASignal::waveform_t∗adm_rtconﬁg_t::lp_coeffs [private]

Lowpass ﬁlter coefﬁcients.

5.33.4.4 std::vector<MHASignal::waveform_t∗>adm_rtconﬁg_t::decomb_coeffs [private]

Decomb-Filter coefﬁcients.

5.33.4.5 std::vector<adm_t ∗>adm_rtconﬁg_t::adms [private]

ADMs.

The documentation for this class was generated from the following ﬁle:

•adm.cpp

5.34 algo_comm_t Struct Reference 201

5.34 algo_comm_t Struct Reference

A reference handle for algorithm communication variables.

Public Attributes

• void ∗handle

AC variable control handle.

• int(∗insert_var )(void ∗, const char ∗,comm_var_t)

• int(∗insert_var_int )(void ∗, const char ∗, int ∗)

• int(∗insert_var_ﬂoat )(void ∗, const char ∗, ﬂoat ∗)

• int(∗remove_var )(void ∗, const char ∗)

Remove an AC variable.

• int(∗remove_ref )(void ∗, void ∗)

Remove all AC variable which refer to address.

• int(∗is_var )(void ∗, const char ∗)

Test if an AC variable exists.

• int(∗get_var )(void ∗, const char ∗,comm_var_t ∗)

Get the variable handle of an AC variable.

• int(∗get_var_int )(void ∗, const char ∗, int ∗)

Get the value of an int AC variable.

• int(∗get_var_ﬂoat )(void ∗, const char ∗, ﬂoat ∗)

Get the value of a ﬂoat AC variable.

• int(∗get_entries )(void ∗, char ∗, unsigned int)

Return a space separated list of all variable names.

• const char ∗(∗get_error )(int)

Convert AC error codes into human readable error messages.

5.34.1 Detailed Description

A reference handle for algorithm communication variables.

This structure contains a coontrol handle and a set of function pointers for sharing variables

within one processing chain. See also section Communication between algorithms (p. 28).

5.34.2 Member Data Documentation

5.34.2.1 algo_comm_t::handle

AC variable control handle.

5.34.2.2 algo_comm_t::insert_var

This function can register a variable to be shared within one chain. If a variable of this name

exists it will be overwritten.

202 CONTENTS

Parameters

hAC handle

nname of variable. May not be empty. Must not contain space character. The name is

copied, therefore it is allowed that the char array pointed to gets invalid after return.

vvariable handle of type comm_var_t (p. 246)

Returns

Error code or zero on success

5.34.2.3 algo_comm_t::insert_var_int

This function can register an int variable to be shared with other algorithms. It behaves similar

to ac.insert_var.

Parameters

hAC handle

nname of variable

vpointer on the variable

Returns

Error code or zero on success

5.34.2.4 algo_comm_t::insert_var_ﬂoat

This function can register a ﬂoat variable to be shared with other algorithms. It behaves similar

to ac.insert_var.

Parameters

hAC handle

nname of variable

vpointer on the variable

Returns

Error code or zero on success

5.34 algo_comm_t Struct Reference 203

5.34.2.5 algo_comm_t::remove_var

Remove an AC variable.

Remove (unregister) an AC variable. After calling this function, the variable is not available to

ac.is_var or ac.get_var. The data pointer is not affected.

Parameters

hAC handle

nname of variable to be removed

Returns

Error code or zero on success

5.34.2.6 algo_comm_t::remove_ref

Remove all AC variable which refer to address.

This function removes all AC variables whos data ﬁeld points to the given address.

Parameters

hAC handle

paddress which should not be referred to any more

Returns

Error code or zero on success

5.34.2.7 algo_comm_t::is_var

Test if an AC variable exists.

This function tests if an AC variable of a given name exists. Use ac.get_var to get information

about the variables type and dimension.

Parameters

hAC handle

nname of variable

204 CONTENTS

Returns

1 if the variable exists, 0 otherwise

5.34.2.8 algo_comm_t::get_var

Get the variable handle of an AC variable.

This function returns the variable handle comm_var_t (p. 246) of a variable of the given name.

If no variable of that name exists, an error code is returned.

Parameters

hAC handle

nname of variable

vpointer to a AC variable object

Returns

Error code or zero on success

5.34.2.9 algo_comm_t::get_var_int

Get the value of an int AC variable.

This function returns the value of an int AC variable of the given name. If no variable exists, the

variable type is mismatching or more than one entry is registered, a corresponding error code

is returned. This is a special version of ac.get_var.

Parameters

hAC handle

nname of variable

vpointer on an int variable to store the result

Returns

Error code or zero on success

5.34.2.10 algo_comm_t::get_var_ﬂoat

Get the value of a ﬂoat AC variable.

This function returns the value of a ﬂoat AC variable of the given name. If no variable exists, the

variable type is mismatching or more than one entry is registered, a corresponding error code

is returned. This is a special version of ac.get_var.

5.34 algo_comm_t Struct Reference 205

Parameters

hAC handle

nname of variable

vpointer on a ﬂoat variable to store the result

Returns

Error code or zero on success

5.34.2.11 algo_comm_t::get_entries

Return a space separated list of all variable names.

This function returns the names of all registered variables, separated by a single space.

Parameters

hAC handle

Return values

ret Character buffer for return value

Parameters

len length of character buffer

Returns

Error code or zero on success. -1: invalid ac handle. -3: not enough room in character

buffer to store all variable names.

5.34.2.12 algo_comm_t::get_error

Convert AC error codes into human readable error messages.

Parameters

eError code

206 CONTENTS

Returns

Error message

The documentation for this struct was generated from the following ﬁles:

•mha.hh

•mha_algo_comm.cpp

5.35 alsa_base_t Class Reference

Inheritance diagram for alsa_base_t:

alsa_base_t

alsa_t< T >

Public Member Functions

•alsa_base_t ()

• virtual ∼alsa_base_t ()=default

• virtual void start ()=0

start puts alsa device in usable state

• virtual void stop ()=0

stop informs alsa device that we do not need any more samples / will not provide any more

samples

• virtual int read (mha_wave_t ∗∗)=0

read audio samples from the device into an internal mha_wave_t (p. 496) buffer, then update

the pointer given as parameter to point to the internal structure.

• virtual int write (mha_wave_t ∗)=0

write audio samples from the given waveform buffer to the sound device.

Public Attributes

• snd_pcm_t ∗pcm

The underlying alsa handle to this sound card.

5.35 alsa_base_t Class Reference 207

5.35.1 Constructor & Destructor Documentation

5.35.1.1 alsa_base_t::alsa_base_t ( ) [inline]

5.35.1.2 virtual alsa_base_t::∼alsa_base_t ( ) [virtual],[default]

5.35.2 Member Function Documentation

5.35.2.1 virtual void alsa_base_t::start ( ) [pure virtual]

start puts alsa device in usable state

Implemented in alsa_t<T>(p. 211).

5.35.2.2 virtual void alsa_base_t::stop ( ) [pure virtual]

stop informs alsa device that we do not need any more samples / will not provide any more

samples

Implemented in alsa_t<T>(p. 212).

5.35.2.3 virtual int alsa_base_t::read (

mha_wave_t ∗∗ )[pure virtual]

read audio samples from the device into an internal mha_wave_t (p. 496) buffer, then update

the pointer given as parameter to point to the internal structure.

Converts sound samples from the integer data type provided by the sound card to ﬂoating-point

values needed by the MHA in the range [-1.0,1.0]

Implemented in alsa_t<T>(p. 212).

5.35.2.4 virtual int alsa_base_t::write (

mha_wave_t ∗)[pure virtual]

write audio samples from the given waveform buffer to the sound device.

converts the ﬂoating point values coming from the MHA to the integer samples required by the

sound card.

Implemented in alsa_t<T>(p. 212).

208 CONTENTS

5.35.3 Member Data Documentation

5.35.3.1 snd_pcm_t∗alsa_base_t::pcm

The underlying alsa handle to this sound card.

The documentation for this class was generated from the following ﬁle:

•MHAIOalsa.cpp

5.36 alsa_dev_par_parser_t Class Reference

Parser variables corresponding to one alsa device.

Inheritance diagram for alsa_dev_par_parser_t:

alsa_dev_par_parser_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•alsa_dev_par_parser_t (snd_pcm_stream_t stream_dir)

Constructor inserts the parser variables into this sub-parser.

Public Attributes

•MHAParser::string_t device

Name of the device in the alsa world, like "hw:0.0", "default", etc.

•MHAParser::int_t nperiods

Number of buffers of fragsize to hold in the alsa buffer.

• snd_pcm_stream_t stream_dir

Remember the direction (capture/playback) of this device.

5.36 alsa_dev_par_parser_t Class Reference 209

Additional Inherited Members

5.36.1 Detailed Description

Parser variables corresponding to one alsa device.

ALSA separates audio capture and audio playback into two different devices that have to be

opened separately. This class encapsulates the parser variables that pertain to one such di-

rection.

5.36.2 Constructor & Destructor Documentation

5.36.2.1 alsa_dev_par_parser_t::alsa_dev_par_parser_t (

snd_pcm_stream_t stream_dir )

Constructor inserts the parser variables into this sub-parser.

Parameters

stream_dir capture or playback

5.36.3 Member Data Documentation

5.36.3.1 MHAParser::string_t alsa_dev_par_parser_t::device

Name of the device in the alsa world, like "hw:0.0", "default", etc.

5.36.3.2 MHAParser::int_t alsa_dev_par_parser_t::nperiods

Number of buffers of fragsize to hold in the alsa buffer.

Usually 2, the minimum possible.

5.36.3.3 snd_pcm_stream_t alsa_dev_par_parser_t::stream_dir

Remember the direction (capture/playback) of this device.

The documentation for this class was generated from the following ﬁle:

•MHAIOalsa.cpp

210 CONTENTS

5.37 alsa_t<T>Class Template Reference

Our representation of one alsa device.

Inheritance diagram for alsa_t<T>:

alsa_t< T >

alsa_base_t

Public Member Functions

•alsa_t (const alsa_dev_par_parser_t &par, unsigned int rate, unsigned int fragsize, un-

signed int channels)

Constructor receives the parameters for this device.

•∼alsa_t ()

Destructor closes the sound device.

• void start () override

start puts alsa device in usable state

• void stop () override

stop informs alsa device that we do not need any more samples / will not provide any more

samples

• int read (mha_wave_t ∗∗) override

read audio samples from the device into an internal mha_wave_t (p. 496) buffer, then update

the pointer given as parameter to point to the internal structure.

• int write (mha_wave_t ∗) override

write audio samples from the given waveform buffer to the sound device.

Private Attributes

• unsigned int channels

• unsigned int fragsize

• T ∗buffer

• std::vector<mha_real_t >frame_data

•MHASignal::waveform_t wave

internal buffer to store sound samples coming from the sound card.

• const mha_real_t gain

• const mha_real_t invgain

• const mha_real_t val_min

• const mha_real_t val_max

• snd_pcm_format_t pcm_format

5.37 alsa_t<T>Class Template Reference 211

Additional Inherited Members

5.37.1 Detailed Description

template<typename T>

class alsa_t<T>

Our representation of one alsa device.

We can start and stop the device, and depending on the direction, read or write samples.

5.37.2 Constructor & Destructor Documentation

5.37.2.1 template<typename T >alsa_t<T>::alsa_t (

const alsa_dev_par_parser_t &par,

unsigned int rate,

unsigned int fragsize,

unsigned int channels )

Constructor receives the parameters for this device.

It opens the sound device using the alsa library and selects the given parameters, but does not

yet start the sound device to perform real I/O.

Parameters

par our parser variable aggregator (containing direction, device name, and number

of periods to place in alsa buffer)

rate sampling rate in Hz

fragsize samples per block per channel

channels number of audio channels to open

5.37.2.2 template<typename T >alsa_t<T>::∼alsa_t ( )

Destructor closes the sound device.

5.37.3 Member Function Documentation

5.37.3.1 template<typename T >void alsa_t<T>::start ( ) [override],[virtual]

start puts alsa device in usable state

Implements alsa_base_t (p. 207).

212 CONTENTS

5.37.3.2 template<typename T >void alsa_t<T>::stop ( ) [override],[virtual]

stop informs alsa device that we do not need any more samples / will not provide any more

samples

Implements alsa_base_t (p. 207).

5.37.3.3 template<typename T >int alsa_t<T>::read (

mha_wave_t ∗∗ s)[override],[virtual]

read audio samples from the device into an internal mha_wave_t (p. 496) buffer, then update

the pointer given as parameter to point to the internal structure.

Converts sound samples from the integer data type provided by the sound card to ﬂoating-point

values needed by the MHA in the range [-1.0,1.0]

Implements alsa_base_t (p. 207).

5.37.3.4 template<typename T >int alsa_t<T>::write (

mha_wave_t ∗s)[override],[virtual]

write audio samples from the given waveform buffer to the sound device.

converts the ﬂoating point values coming from the MHA to the integer samples required by the

sound card.

Implements alsa_base_t (p. 207).

5.37.4 Member Data Documentation

5.37.4.1 template<typename T >unsigned int alsa_t<T>::channels [private]

5.37.4.2 template<typename T >unsigned int alsa_t<T>::fragsize [private]

5.37.4.3 template<typename T >T∗alsa_t<T>::buffer [private]

5.37.4.4 template<typename T >std::vector<mha_real_t>alsa_t<T>::frame_data [private]

5.37.4.5 template<typename T >MHASignal::waveform_t alsa_t<T>::wave [private]

internal buffer to store sound samples coming from the sound card.

5.38 altplugs_t Class Reference 213

5.37.4.6 template<typename T >const mha_real_t alsa_t<T>::gain [private]

5.37.4.7 template<typename T >const mha_real_t alsa_t<T>::invgain [private]

5.37.4.8 template<typename T >const mha_real_t alsa_t<T>::val_min [private]

5.37.4.9 template<typename T >const mha_real_t alsa_t<T>::val_max [private]

5.37.4.10 template<typename T >snd_pcm_format_t alsa_t<T>::pcm_format [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOalsa.cpp

5.38 altplugs_t Class Reference

Inheritance diagram for altplugs_t:

altplugs_t

MHAPlugin::plugin_t

< MHAWindow::fun_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< MHAWindow::fun_t >

Public Member Functions

•altplugs_t (algo_comm_t iac, const char ∗chain, const char ∗algo)

• void prepare (mhaconﬁg_t &)

• void release ()

• void process (mha_wave_t ∗,mha_wave_t ∗∗)

• void process (mha_spec_t ∗,mha_wave_t ∗∗)

• void process (mha_wave_t ∗,mha_spec_t ∗∗)

• void process (mha_spec_t ∗,mha_spec_t ∗∗)

• virtual std::string parse (const std::string &arg)

• virtual void parse (const char ∗a1, char ∗a2, unsigned int a3)

214 CONTENTS

Private Member Functions

• void event_set_plugs ()

• void event_add_plug ()

• void event_delete_plug ()

• void event_select_plug ()

• void update_selector_list ()

• void update_ramplen ()

• void proc_ramp (mha_wave_t ∗s)

Private Attributes

•MHAParser::bool_t use_own_ac

•MHAParser::vstring_t parser_plugs

•MHAParser::string_t add_plug

•MHAParser::string_t delete_plug

•MHAParser::ﬂoat_t ramplen

•MHAParser::kw_t select_plug

•MHAParser::parser_t current

•MHAParser::vstring_mon_t nondefault_labels

• std::vector<mhaplug_cfg_t ∗>plugs

•mhaplug_cfg_t ∗selected_plug

•MHAEvents::patchbay_t<altplugs_t >patchbay

•MHASignal::waveform_t ∗fallback_wave

•MHASignal::spectrum_t ∗fallback_spec

•mhaconﬁg_t cﬁn

•mhaconﬁg_t cfout

• bool prepared

• bool added_via_plugs

• unsigned int ramp_counter

• unsigned int ramp_len

Additional Inherited Members

5.38.1 Constructor & Destructor Documentation

5.38.1.1 altplugs_t::altplugs_t (

algo_comm_t iac,

const char ∗chain,

const char ∗algo )

5.38.2 Member Function Documentation

5.38.2.1 void altplugs_t::prepare (

mhaconﬁg_t &cf )[virtual]

Implements MHAPlugin::plugin_t<MHAWindow::fun_t >(p. 730).

5.38 altplugs_t Class Reference 215

5.38.2.2 void altplugs_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<MHAWindow::fun_t >(p. 731).

5.38.2.3 void altplugs_t::process (

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )

5.38.2.4 void altplugs_t::process (

mha_spec_t ∗sIn,

mha_wave_t ∗∗ sOut )

5.38.2.5 void altplugs_t::process (

mha_wave_t ∗sIn,

mha_spec_t ∗∗ sOut )

5.38.2.6 void altplugs_t::process (

mha_spec_t ∗sIn,

mha_spec_t ∗∗ sOut )

5.38.2.7 std::string altplugs_t::parse (

const std::string & arg )[virtual]

Reimplemented from MHAParser::base_t (p. 637).

5.38.2.8 virtual void altplugs_t::parse (

const char ∗a1,

char ∗a2,

unsigned int a3 )[inline],[virtual]

Reimplemented from MHAParser::base_t (p. 638).

5.38.2.9 void altplugs_t::event_set_plugs ( ) [private]

5.38.2.10 void altplugs_t::event_add_plug ( ) [private]

5.38.2.11 void altplugs_t::event_delete_plug ( ) [private]

5.38.2.12 void altplugs_t::event_select_plug ( ) [private]

5.38.2.13 void altplugs_t::update_selector_list ( ) [private]

5.38.2.14 void altplugs_t::update_ramplen ( ) [private]

216 CONTENTS

5.38.2.15 void altplugs_t::proc_ramp (

mha_wave_t ∗s)[private]

5.38.3 Member Data Documentation

5.38.3.1 MHAParser::bool_t altplugs_t::use_own_ac [private]

5.38.3.2 MHAParser::vstring_t altplugs_t::parser_plugs [private]

5.38.3.3 MHAParser::string_t altplugs_t::add_plug [private]

5.38.3.4 MHAParser::string_t altplugs_t::delete_plug [private]

5.38.3.5 MHAParser::ﬂoat_t altplugs_t::ramplen [private]

5.38.3.6 MHAParser::kw_t altplugs_t::select_plug [private]

5.38.3.7 MHAParser::parser_t altplugs_t::current [private]

5.38.3.8 MHAParser::vstring_mon_t altplugs_t::nondefault_labels [private]

5.38.3.9 std::vector<mhaplug_cfg_t∗>altplugs_t::plugs [private]

5.38.3.10 mhaplug_cfg_t∗altplugs_t::selected_plug [private]

5.38.3.11 MHAEvents::patchbay_t<altplugs_t>altplugs_t::patchbay [private]

5.38.3.12 MHASignal::waveform_t∗altplugs_t::fallback_wave [private]

5.38.3.13 MHASignal::spectrum_t∗altplugs_t::fallback_spec [private]

5.38.3.14 mhaconﬁg_t altplugs_t::cﬁn [private]

5.38.3.15 mhaconﬁg_t altplugs_t::cfout [private]

5.38.3.16 bool altplugs_t::prepared [private]

5.38.3.17 bool altplugs_t::added_via_plugs [private]

5.38.3.18 unsigned int altplugs_t::ramp_counter [private]

5.38.3.19 unsigned int altplugs_t::ramp_len [private]

The documentation for this class was generated from the following ﬁle:

•altplugs.cpp

5.39 analysepath_t Class Reference 217

5.39 analysepath_t Class Reference

Public Member Functions

•analysepath_t (unsigned int nchannels_in, unsigned int outer_fragsize, unsigned int

inner_fragsize, int priority,MHAProc_wave2wave_t inner_proc_wave2wave, MHA←-

Proc_wave2spec_t inner_proc_wave2spec, void ∗ilibdata, algo_comm_t outer_ac,

const MHA_AC::acspace2matrix_t &acspace_template, mha_domain_t inner_out_←-

domain, unsigned int ﬁfo_len_blocks)

• virtual ∼analysepath_t ()

• void rt_process (mha_wave_t ∗)

• virtual int svc ()

Private Attributes

•MHAProc_wave2wave_t inner_process_wave2wave

•MHAProc_wave2spec_t inner_process_wave2spec

•MHASignal::waveform_t inner_input

• void ∗libdata

•mha_ﬁfo_t<mha_real_t >wave_ﬁfo

•mha_ﬁfo_t<MHA_AC::acspace2matrix_t >ac_ﬁfo

•MHA_AC::acspace2matrix_t inner_ac_copy

•MHA_AC::acspace2matrix_t outer_ac_copy

•algo_comm_t outer_ac

•mha_domain_t inner_out_domain

•MHA_Error inner_error

• bool has_inner_error

• bool ﬂag_terminate_inner_thread

• int input_to_process

• pthread_mutex_t ProcessMutex

• pthread_attr_t attr

• struct sched_param priority

• int scheduler

• pthread_t thread

• pthread_cond_t cond_to_process

5.39.1 Constructor & Destructor Documentation

5.39.1.1 analysepath_t::analysepath_t (

unsigned int nchannels_in,

unsigned int outer_fragsize,

unsigned int inner_fragsize,

int priority,

MHAProc_wave2wave_t inner_proc_wave2wave,

MHAProc_wave2spec_t inner_proc_wave2spec,

void ∗ilibdata,

algo_comm_t outer_ac,

const MHA_AC::acspace2matrix_t &acspace_template,

mha_domain_t inner_out_domain,

unsigned int ﬁfo_len_blocks )

218 CONTENTS

5.39.1.2 analysepath_t::∼analysepath_t ( ) [virtual]

5.39.2 Member Function Documentation

5.39.2.1 void analysepath_t::rt_process (

mha_wave_t ∗outer_input )

5.39.2.2 int analysepath_t::svc ( ) [virtual]

5.39.3 Member Data Documentation

5.39.3.1 MHAProc_wave2wave_t analysepath_t::inner_process_wave2wave [private]

5.39.3.2 MHAProc_wave2spec_t analysepath_t::inner_process_wave2spec [private]

5.39.3.3 MHASignal::waveform_t analysepath_t::inner_input [private]

5.39.3.4 void∗analysepath_t::libdata [private]

5.39.3.5 mha_ﬁfo_t<mha_real_t>analysepath_t::wave_ﬁfo [private]

5.39.3.6 mha_ﬁfo_t<MHA_AC::acspace2matrix_t>analysepath_t::ac_ﬁfo [private]

5.39.3.7 MHA_AC::acspace2matrix_t analysepath_t::inner_ac_copy [private]

5.39.3.8 MHA_AC::acspace2matrix_t analysepath_t::outer_ac_copy [private]

5.39.3.9 algo_comm_t analysepath_t::outer_ac [private]

5.39.3.10 mha_domain_t analysepath_t::inner_out_domain [private]

5.39.3.11 MHA_Error analysepath_t::inner_error [private]

5.39.3.12 bool analysepath_t::has_inner_error [private]

5.39.3.13 bool analysepath_t::ﬂag_terminate_inner_thread [private]

5.39.3.14 int analysepath_t::input_to_process [private]

5.39.3.15 pthread_mutex_t analysepath_t::ProcessMutex [private]

5.39.3.16 pthread_attr_t analysepath_t::attr [private]

5.39.3.17 struct sched_param analysepath_t::priority [private]

5.39.3.18 int analysepath_t::scheduler [private]

5.39.3.19 pthread_t analysepath_t::thread [private]

5.39.3.20 pthread_cond_t analysepath_t::cond_to_process [private]

The documentation for this class was generated from the following ﬁle:

•analysispath.cpp

5.40 analysispath_if_t Class Reference 219

5.40 analysispath_if_t Class Reference

Inheritance diagram for analysispath_if_t:

analysispath_if_t

MHAPlugin::plugin_t

< analysepath_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< analysepath_t >

Public Member Functions

•analysispath_if_t (algo_comm_t, std::string, std::string)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

•∼analysispath_if_t ()

Private Member Functions

• void loadlib ()

Private Attributes

•MHAEvents::patchbay_t<analysispath_if_t >patchbay

•MHAParser::string_t libname

•MHAParser::int_t fragsize

•MHAParser::int_t ﬁfolen

•MHAParser::int_t priority

•MHAParser::vstring_t vars

•plug_t ∗plug

• std::string chain

• std::string algo

•MHA_AC::acspace2matrix_t ∗acspace_template

220 CONTENTS

Additional Inherited Members

5.40.1 Constructor & Destructor Documentation

5.40.1.1 analysispath_if_t::analysispath_if_t (

algo_comm_t iac,

std::string th,

std::string al )

5.40.1.2 analysispath_if_t::∼analysispath_if_t ( )

5.40.2 Member Function Documentation

5.40.2.1 mha_wave_t ∗analysispath_if_t::process (

mha_wave_t ∗s)

5.40.2.2 void analysispath_if_t::prepare (

mhaconﬁg_t &conf )[virtual]

Implements MHAPlugin::plugin_t<analysepath_t >(p. 730).

5.40.2.3 void analysispath_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<analysepath_t >(p. 731).

5.40.2.4 void analysispath_if_t::loadlib ( ) [private]

5.40.3 Member Data Documentation

5.40.3.1 MHAEvents::patchbay_t<analysispath_if_t >analysispath_if_t::patchbay

[private]

5.40.3.2 MHAParser::string_t analysispath_if_t::libname [private]

5.40.3.3 MHAParser::int_t analysispath_if_t::fragsize [private]

5.40.3.4 MHAParser::int_t analysispath_if_t::ﬁfolen [private]

5.40.3.5 MHAParser::int_t analysispath_if_t::priority [private]

5.40.3.6 MHAParser::vstring_t analysispath_if_t::vars [private]

5.40.3.7 plug_t∗analysispath_if_t::plug [private]

5.40.3.8 std::string analysispath_if_t::chain [private]

5.40.3.9 std::string analysispath_if_t::algo [private]

5.40.3.10 MHA_AC::acspace2matrix_t∗analysispath_if_t::acspace_template [private]

The documentation for this class was generated from the following ﬁle:

•analysispath.cpp

5.41 AuditoryProﬁle::fmap_t Class Reference 221

5.41 AuditoryProﬁle::fmap_t Class Reference

A class to store frequency dependent data (e.g., HTL and UCL).

Inherits map<mha_real_t, mha_real_t >.

Public Member Functions

• std::vector<mha_real_t >get_frequencies () const

Return conﬁgured frequencies.

• std::vector<mha_real_t >get_values () const

Return stored values corresponding to the frequencies.

• bool isempty () const

5.41.1 Detailed Description

A class to store frequency dependent data (e.g., HTL and UCL).

5.41.2 Member Function Documentation

5.41.2.1 std::vector<mha_real_t >AuditoryProﬁle::fmap_t::get_frequencies ( ) const

Return conﬁgured frequencies.

5.41.2.2 std::vector<mha_real_t >AuditoryProﬁle::fmap_t::get_values ( ) const

Return stored values corresponding to the frequencies.

5.41.2.3 bool AuditoryProﬁle::fmap_t::isempty ( ) const [inline]

The documentation for this class was generated from the following ﬁles:

•auditory_proﬁle.h

•auditory_proﬁle.cpp

222 CONTENTS

5.42 AuditoryProﬁle::parser_t Class Reference

Class to make the auditory proﬁle accessible through the parser interface.

Inheritance diagram for AuditoryProﬁle::parser_t:

AuditoryProfile::parser_t

MHAParser::parser_t

MHAParser::base_t

Classes

• class ear_t

• class fmap_t

Public Member Functions

•parser_t ()

•AuditoryProﬁle::proﬁle_t get_current_proﬁle ()

Private Attributes

•AuditoryProﬁle::parser_t::ear_t L

•AuditoryProﬁle::parser_t::ear_t R

Additional Inherited Members

5.42.1 Detailed Description

Class to make the auditory proﬁle accessible through the parser interface.

5.43 AuditoryProﬁle::parser_t::ear_t Class Reference 223

5.42.2 Constructor & Destructor Documentation

5.42.2.1 AuditoryProﬁle::parser_t::parser_t ( )

5.42.3 Member Function Documentation

5.42.3.1 AuditoryProﬁle::proﬁle_t AuditoryProﬁle::parser_t::get_current_proﬁle ( )

5.42.4 Member Data Documentation

5.42.4.1 AuditoryProﬁle::parser_t::ear_t AuditoryProﬁle::parser_t::L [private]

5.42.4.2 AuditoryProﬁle::parser_t::ear_t AuditoryProﬁle::parser_t::R [private]

The documentation for this class was generated from the following ﬁles:

•auditory_proﬁle.h

•auditory_proﬁle.cpp

5.43 AuditoryProﬁle::parser_t::ear_t Class Reference

Inheritance diagram for AuditoryProﬁle::parser_t::ear_t:

AuditoryProfile::parser

_t::ear_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•ear_t ()

•AuditoryProﬁle::proﬁle_t::ear_t get_ear () const

224 CONTENTS

Private Attributes

•AuditoryProﬁle::parser_t::fmap_t HTL

•AuditoryProﬁle::parser_t::fmap_t UCL

Additional Inherited Members

5.43.1 Constructor & Destructor Documentation

5.43.1.1 AuditoryProﬁle::parser_t::ear_t::ear_t ( )

5.43.2 Member Function Documentation

5.43.2.1 AuditoryProﬁle::proﬁle_t::ear_t AuditoryProﬁle::parser_t::ear_t::get_ear ( ) const

5.43.3 Member Data Documentation

5.43.3.1 AuditoryProﬁle::parser_t::fmap_t AuditoryProﬁle::parser_t::ear_t::HTL [private]

5.43.3.2 AuditoryProﬁle::parser_t::fmap_t AuditoryProﬁle::parser_t::ear_t::UCL [private]

The documentation for this class was generated from the following ﬁles:

•auditory_proﬁle.h

•auditory_proﬁle.cpp

5.44 AuditoryProﬁle::parser_t::fmap_t Class Reference

Inheritance diagram for AuditoryProﬁle::parser_t::fmap_t:

AuditoryProfile::parser

_t::fmap_t

MHAParser::parser_t

MHAParser::base_t

5.44 AuditoryProﬁle::parser_t::fmap_t Class Reference 225

Public Member Functions

•fmap_t (const std::string &name, const std::string &help)

•AuditoryProﬁle::fmap_t get_fmap () const

Private Member Functions

• void validate ()

Private Attributes

•MHAEvents::patchbay_t<AuditoryProﬁle::parser_t::fmap_t >patchbay

•MHAParser::vﬂoat_t f

•MHAParser::vﬂoat_t value

• std::string name_

Additional Inherited Members

5.44.1 Constructor & Destructor Documentation

5.44.1.1 AuditoryProﬁle::parser_t::fmap_t::fmap_t (

const std::string & name,

const std::string & help )

5.44.2 Member Function Documentation

5.44.2.1 AuditoryProﬁle::fmap_t AuditoryProﬁle::parser_t::fmap_t::get_fmap ( ) const

5.44.2.2 void AuditoryProﬁle::parser_t::fmap_t::validate ( ) [private]

5.44.3 Member Data Documentation

5.44.3.1 MHAEvents::patchbay_t<AuditoryProﬁle::parser_t::fmap_t>

AuditoryProﬁle::parser_t::fmap_t::patchbay [private]

5.44.3.2 MHAParser::vﬂoat_t AuditoryProﬁle::parser_t::fmap_t::f [private]

5.44.3.3 MHAParser::vﬂoat_t AuditoryProﬁle::parser_t::fmap_t::value [private]

5.44.3.4 std::string AuditoryProﬁle::parser_t::fmap_t::name_ [private]

The documentation for this class was generated from the following ﬁles:

•auditory_proﬁle.h

•auditory_proﬁle.cpp

226 CONTENTS

5.45 AuditoryProﬁle::proﬁle_t Class Reference

The Auditory Proﬁle class.

Classes

• class ear_t

Class for ear-dependent parameters, e.g., audiograms or unilateral loudness scaling.

Public Member Functions

•AuditoryProﬁle::proﬁle_t::ear_t get_ear (unsigned int channel) const

Return ear information of channel number.

Public Attributes

•AuditoryProﬁle::proﬁle_t::ear_t L

Left ear data.

•AuditoryProﬁle::proﬁle_t::ear_t R

Right ear data.

5.45.1 Detailed Description

The Auditory Proﬁle class.

See deﬁnition of auditory proﬁle

Todo Give more documentation; implement all parts of the auditory proﬁle.

Currently only the audiogram data is stored.

5.45.2 Member Function Documentation

5.45.2.1 AuditoryProﬁle::proﬁle_t::ear_t AuditoryProﬁle::proﬁle_t::get_ear (

unsigned int channel ) const [inline]

Return ear information of channel number.

5.46 AuditoryProﬁle::proﬁle_t::ear_t Class Reference 227

5.45.3 Member Data Documentation

5.45.3.1 AuditoryProﬁle::proﬁle_t::ear_t AuditoryProﬁle::proﬁle_t::L

Left ear data.

5.45.3.2 AuditoryProﬁle::proﬁle_t::ear_t AuditoryProﬁle::proﬁle_t::R

Right ear data.

The documentation for this class was generated from the following ﬁle:

•auditory_proﬁle.h

5.46 AuditoryProﬁle::proﬁle_t::ear_t Class Reference

Class for ear-dependent parameters, e.g., audiograms or unilateral loudness scaling.

Public Member Functions

• void convert_empty2normal ()

Public Attributes

•AuditoryProﬁle::fmap_t HTL

•AuditoryProﬁle::fmap_t UCL

5.46.1 Detailed Description

Class for ear-dependent parameters, e.g., audiograms or unilateral loudness scaling.

5.46.2 Member Function Documentation

5.46.2.1 void AuditoryProﬁle::proﬁle_t::ear_t::convert_empty2normal ( )

5.46.3 Member Data Documentation

5.46.3.1 AuditoryProﬁle::fmap_t AuditoryProﬁle::proﬁle_t::ear_t::HTL

5.46.3.2 AuditoryProﬁle::fmap_t AuditoryProﬁle::proﬁle_t::ear_t::UCL

The documentation for this class was generated from the following ﬁles:

•auditory_proﬁle.h

•auditory_proﬁle.cpp

228 CONTENTS

5.47 bbcalib_interface_t Class Reference

Inheritance diagram for bbcalib_interface_t:

bbcalib_interface_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•bbcalib_interface_t (const algo_comm_t &, const std::string &, const std::string &)

•∼bbcalib_interface_t ()

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Attributes

•calibrator_t calib_in

•calibrator_t calib_out

•MHAParser::mhapluginloader_t plugloader

5.48 calibrator_runtime_layer_t Class Reference 229

Additional Inherited Members

5.47.1 Constructor & Destructor Documentation

5.47.1.1 bbcalib_interface_t::bbcalib_interface_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.47.1.2 bbcalib_interface_t::∼bbcalib_interface_t ( )

5.47.2 Member Function Documentation

5.47.2.1 mha_wave_t ∗bbcalib_interface_t::process (

mha_wave_t ∗s)

5.47.2.2 void bbcalib_interface_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<int >(p. 730).

5.47.2.3 void bbcalib_interface_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.47.3 Member Data Documentation

5.47.3.1 calibrator_t bbcalib_interface_t::calib_in [private]

5.47.3.2 calibrator_t bbcalib_interface_t::calib_out [private]

5.47.3.3 MHAParser::mhapluginloader_t bbcalib_interface_t::plugloader [private]

The documentation for this class was generated from the following ﬁle:

•transducers.cpp

5.48 calibrator_runtime_layer_t Class Reference

Public Member Functions

•calibrator_runtime_layer_t (bool is_input, const mhaconﬁg_t &tf, calibrator_←-

variables_t &vars)

•mha_real_t process (mha_wave_t ∗∗)

230 CONTENTS

Static Private Member Functions

• static unsigned int ﬁrﬁrlen (const std::vector<std::vector<ﬂoat > > &)

• static unsigned int ﬁrﬁr2fftlen (unsigned int, const std::vector<std::vector<ﬂoat > > &)

Private Attributes

•MHAFilter::fftﬁlter_t ﬁr

•MHASignal::quantizer_t quant

•MHASignal::waveform_t gain

•softclipper_t softclip

• bool b_is_input

• bool b_use_ﬁr

• bool b_use_clipping

•MHASignal::loop_wavefragment_t speechnoise

•MHASignal::loop_wavefragment_t::playback_mode_t pmode

5.48.1 Constructor & Destructor Documentation

5.48.1.1 calibrator_runtime_layer_t::calibrator_runtime_layer_t (

bool is_input,

const mhaconﬁg_t &tf,

calibrator_variables_t &vars )

5.48.2 Member Function Documentation

5.48.2.1 mha_real_t calibrator_runtime_layer_t::process (

mha_wave_t ∗∗ s)

5.48.2.2 unsigned int calibrator_runtime_layer_t::ﬁrﬁrlen (

const std::vector<std::vector<ﬂoat > > &ﬁr )[static],[private]

5.48.2.3 unsigned int calibrator_runtime_layer_t::ﬁrﬁr2fftlen (

unsigned int fragsize,

const std::vector<std::vector<ﬂoat > > &ﬁr )[static],[private]

5.48.3 Member Data Documentation

5.48.3.1 MHAFilter::fftﬁlter_t calibrator_runtime_layer_t::ﬁr [private]

5.48.3.2 MHASignal::quantizer_t calibrator_runtime_layer_t::quant [private]

5.48.3.3 MHASignal::waveform_t calibrator_runtime_layer_t::gain [private]

5.49 calibrator_t Class Reference 231

5.48.3.4 softclipper_t calibrator_runtime_layer_t::softclip [private]

5.48.3.5 bool calibrator_runtime_layer_t::b_is_input [private]

5.48.3.6 bool calibrator_runtime_layer_t::b_use_ﬁr [private]

5.48.3.7 bool calibrator_runtime_layer_t::b_use_clipping [private]

5.48.3.8 MHASignal::loop_wavefragment_t calibrator_runtime_layer_t::speechnoise

[private]

5.48.3.9 MHASignal::loop_wavefragment_t::playback_mode_t calibrator_runtime_layer_t←-

::pmode [private]

The documentation for this class was generated from the following ﬁle:

•transducers.cpp

5.49 calibrator_t Class Reference

Inheritance diagram for calibrator_t:

calibrator_t

rtcalibrator

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< runtime_cfg_t >

rmslevelmeter

232 CONTENTS

Public Member Functions

•calibrator_t (algo_comm_t, bool is_input)

• void prepare (mhaconﬁg_t &tf)

• void release ()

•mha_wave_t ∗process (mha_wave_t ∗s)

Private Member Functions

• void update ()

• void update_tau_level ()

• void read_levels ()

Private Attributes

• bool b_is_input

•MHAEvents::patchbay_t<calibrator_t >patchbay

•calibrator_variables_t vars

• bool prepared

Additional Inherited Members

5.49.1 Constructor & Destructor Documentation

5.49.1.1 calibrator_t::calibrator_t (

algo_comm_t iac,

bool is_input )

5.49.2 Member Function Documentation

5.49.2.1 void calibrator_t::prepare (

mhaconﬁg_t &tf )[inline],[virtual]

Implements MHAPlugin::plugin_t<runtime_cfg_t >(p. 730).

5.49.2.2 void calibrator_t::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<runtime_cfg_t >(p. 731).

5.50 calibrator_variables_t Class Reference 233

5.49.2.3 mha_wave_t ∗calibrator_t::process (

mha_wave_t ∗s)

5.49.2.4 void calibrator_t::update ( ) [private]

5.49.2.5 void calibrator_t::update_tau_level ( ) [private]

5.49.2.6 void calibrator_t::read_levels ( ) [private]

5.49.3 Member Data Documentation

5.49.3.1 bool calibrator_t::b_is_input [private]

5.49.3.2 MHAEvents::patchbay_t<calibrator_t>calibrator_t::patchbay [private]

5.49.3.3 calibrator_variables_t calibrator_t::vars [private]

5.49.3.4 bool calibrator_t::prepared [private]

The documentation for this class was generated from the following ﬁle:

•transducers.cpp

5.50 calibrator_variables_t Class Reference

Public Member Functions

•calibrator_variables_t (bool is_input, MHAParser::parser_t &parent)

Public Attributes

•MHAParser::vﬂoat_t peaklevel

•MHAParser::mﬂoat_t ﬁr

•MHAParser::int_t nbits

•MHAParser::ﬂoat_t tau_level

•MHAParser::kw_t spnoise_mode

•MHAParser::vint_t spnoise_channels

•MHAParser::ﬂoat_t spnoise_level

•MHAParser::vﬂoat_mon_t rmslevel

•MHAParser::parser_t spnoise_parser

•MHAParser::ﬂoat_mon_t srate

•MHAParser::int_mon_t fragsize

•MHAParser::int_mon_t num_channels

•MHAParser::parser_t conﬁg_parser

•softclipper_variables_t softclip

•MHAParser::bool_t do_clipping

234 CONTENTS

5.50.1 Constructor & Destructor Documentation

5.50.1.1 calibrator_variables_t::calibrator_variables_t (

bool is_input,

MHAParser::parser_t &parent )

5.50.2 Member Data Documentation

5.50.2.1 MHAParser::vﬂoat_t calibrator_variables_t::peaklevel

5.50.2.2 MHAParser::mﬂoat_t calibrator_variables_t::ﬁr

5.50.2.3 MHAParser::int_t calibrator_variables_t::nbits

5.50.2.4 MHAParser::ﬂoat_t calibrator_variables_t::tau_level

5.50.2.5 MHAParser::kw_t calibrator_variables_t::spnoise_mode

5.50.2.6 MHAParser::vint_t calibrator_variables_t::spnoise_channels

5.50.2.7 MHAParser::ﬂoat_t calibrator_variables_t::spnoise_level

5.50.2.8 MHAParser::vﬂoat_mon_t calibrator_variables_t::rmslevel

5.50.2.9 MHAParser::parser_t calibrator_variables_t::spnoise_parser

5.50.2.10 MHAParser::ﬂoat_mon_t calibrator_variables_t::srate

5.50.2.11 MHAParser::int_mon_t calibrator_variables_t::fragsize

5.50.2.12 MHAParser::int_mon_t calibrator_variables_t::num_channels

5.50.2.13 MHAParser::parser_t calibrator_variables_t::conﬁg_parser

5.50.2.14 softclipper_variables_t calibrator_variables_t::softclip

5.50.2.15 MHAParser::bool_t calibrator_variables_t::do_clipping

The documentation for this class was generated from the following ﬁle:

•transducers.cpp

5.51 cfg_t Class Reference 235

5.51 cfg_t Class Reference

Public Member Functions

•cfg_t (unsigned int, unsigned int)

•cfg_t (mhaconﬁg_t chcfg, mha_real_t newlev, bool replace, mha_real_t len)

• void process (mha_wave_t ∗)

• void process (mha_spec_t ∗)

•cfg_t (mha_real_t tau_attack, mha_real_t tau_decay, unsigned int nch, mha_real_←-

tstart_limit, mha_real_t slope_db, mha_real_t fs)

Public Attributes

• unsigned int channel

•mha_real_t start_lin

•mha_real_t alpha

•MHAFilter::o1ﬂt_lowpass_t attack

•MHAFilter::o1ﬂt_maxtrack_t decay

Private Attributes

•mha_real_t gain_wave_

•mha_real_t gain_spec_

• bool replace_

• bool use_frozen_

•MHASignal::waveform_t frozen_noise_

• unsigned int pos

5.51.1 Constructor & Destructor Documentation

5.51.1.1 cfg_t::cfg_t (

unsigned int ichannel,

unsigned int numchannels )

5.51.1.2 cfg_t::cfg_t (

mhaconﬁg_t chcfg,

mha_real_t newlev,

bool replace,

mha_real_t len )

5.51.1.3 cfg_t::cfg_t (

mha_real_t tau_attack,

mha_real_t tau_decay,

unsigned int nch,

mha_real_t start_limit,

mha_real_t slope_db,

mha_real_t fs )

236 CONTENTS

5.51.2 Member Function Documentation

5.51.2.1 void cfg_t::process (

mha_wave_t ∗s)[inline]

5.51.2.2 void cfg_t::process (

mha_spec_t ∗s)[inline]

5.51.3 Member Data Documentation

5.51.3.1 unsigned int cfg_t::channel

5.51.3.2 mha_real_t cfg_t::gain_wave_ [private]

5.51.3.3 mha_real_t cfg_t::gain_spec_ [private]

5.51.3.4 bool cfg_t::replace_ [private]

5.51.3.5 bool cfg_t::use_frozen_ [private]

5.51.3.6 MHASignal::waveform_t cfg_t::frozen_noise_ [private]

5.51.3.7 unsigned int cfg_t::pos [private]

5.51.3.8 mha_real_t cfg_t::start_lin

5.51.3.9 mha_real_t cfg_t::alpha

5.51.3.10 MHAFilter::o1ﬂt_lowpass_t cfg_t::attack

5.51.3.11 MHAFilter::o1ﬂt_maxtrack_t cfg_t::decay

The documentation for this class was generated from the following ﬁles:

•example6.cpp

•noise.cpp

•softclip.cpp

5.52 coherence::cohﬂt_if_t Class Reference 237

5.52 coherence::cohﬂt_if_t Class Reference

Inheritance diagram for coherence::cohﬂt_if_t:

coherence::cohflt_if_t

MHAPlugin::plugin_t

< cohflt_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< cohflt_t >

Public Member Functions

•cohﬂt_if_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare (mhaconﬁg_t &)

• void release ()

•mha_spec_t ∗process (mha_spec_t ∗)

Private Member Functions

• void update ()

Private Attributes

•MHAEvents::patchbay_t<cohﬂt_if_t >patchbay

•vars_t vars

• const std::string algo

238 CONTENTS

Additional Inherited Members

5.52.1 Constructor & Destructor Documentation

5.52.1.1 coherence::cohﬂt_if_t::cohﬂt_if_t (

const algo_comm_t &ac,

const std::string & th,

const std::string & al )

5.52.2 Member Function Documentation

5.52.2.1 void coherence::cohﬂt_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<cohﬂt_t >(p. 730).

5.52.2.2 void coherence::cohﬂt_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<cohﬂt_t >(p. 731).

5.52.2.3 mha_spec_t ∗coherence::cohﬂt_if_t::process (

mha_spec_t ∗s)

5.52.2.4 void coherence::cohﬂt_if_t::update ( ) [private]

5.52.3 Member Data Documentation

5.52.3.1 MHAEvents::patchbay_t<cohﬂt_if_t>coherence::cohﬂt_if_t::patchbay [private]

5.52.3.2 vars_t coherence::cohﬂt_if_t::vars [private]

5.52.3.3 const std::string coherence::cohﬂt_if_t::algo [private]

The documentation for this class was generated from the following ﬁle:

•coherence.cpp

5.53 coherence::cohﬂt_t Class Reference 239

5.53 coherence::cohﬂt_t Class Reference

Inheritance diagram for coherence::cohﬂt_t:

coherence::cohflt_t

MHAOvlFilter::fftfb_t

MHAOvlFilter::fspacing_t MHASignal::waveform_t

mha_wave_t

mhaconfig_t

Public Member Functions

•cohﬂt_t (vars_t &v, const mhaconﬁg_t &icf, algo_comm_t iac, const std::string &name)

•mha_spec_t ∗process (mha_spec_t ∗)

• void insert ()

Private Attributes

• unsigned int nbands

• bool avg_ipd

•mha_complex_t cg

• ﬂoat g

• ﬂoat c_scale

• ﬂoat c_min

•MHASignal::waveform_t alpha

• ﬂoat limit

•MHAFilter::o1ﬂt_lowpass_t lp1r

•MHAFilter::o1ﬂt_lowpass_t lp1i

•MHA_AC::spectrum_t coh_c

•MHA_AC::waveform_t coh_rlp

•MHASignal::waveform_t gain

•MHASignal::delay_wave_t gain_delay

240 CONTENTS

•MHASignal::spectrum_t s_out

• bool bInvert

•MHAFilter::o1ﬂt_lowpass_t lp1ltg

• bool b_ltg

• std::vector<ﬂoat >staticgain

Additional Inherited Members

5.53.1 Constructor & Destructor Documentation

5.53.1.1 coherence::cohﬂt_t::cohﬂt_t (

vars_t &v,

const mhaconﬁg_t &icf,

algo_comm_t iac,

const std::string & name )

5.53.2 Member Function Documentation

5.53.2.1 mha_spec_t ∗coherence::cohﬂt_t::process (

mha_spec_t ∗s)

5.53.2.2 void coherence::cohﬂt_t::insert ( )

5.53.3 Member Data Documentation

5.53.3.1 unsigned int coherence::cohﬂt_t::nbands [private]

5.53.3.2 bool coherence::cohﬂt_t::avg_ipd [private]

5.53.3.3 mha_complex_t coherence::cohﬂt_t::cg [private]

5.53.3.4 ﬂoat coherence::cohﬂt_t::g [private]

5.53.3.5 ﬂoat coherence::cohﬂt_t::c_scale [private]

5.53.3.6 ﬂoat coherence::cohﬂt_t::c_min [private]

5.53.3.7 MHASignal::waveform_t coherence::cohﬂt_t::alpha [private]

5.53.3.8 ﬂoat coherence::cohﬂt_t::limit [private]

5.53.3.9 MHAFilter::o1ﬂt_lowpass_t coherence::cohﬂt_t::lp1r [private]

5.54 coherence::vars_t Class Reference 241

5.53.3.10 MHAFilter::o1ﬂt_lowpass_t coherence::cohﬂt_t::lp1i [private]

5.53.3.11 MHA_AC::spectrum_t coherence::cohﬂt_t::coh_c [private]

5.53.3.12 MHA_AC::waveform_t coherence::cohﬂt_t::coh_rlp [private]

5.53.3.13 MHASignal::waveform_t coherence::cohﬂt_t::gain [private]

5.53.3.14 MHASignal::delay_wave_t coherence::cohﬂt_t::gain_delay [private]

5.53.3.15 MHASignal::spectrum_t coherence::cohﬂt_t::s_out [private]

5.53.3.16 bool coherence::cohﬂt_t::bInvert [private]

5.53.3.17 MHAFilter::o1ﬂt_lowpass_t coherence::cohﬂt_t::lp1ltg [private]

5.53.3.18 bool coherence::cohﬂt_t::b_ltg [private]

5.53.3.19 std::vector<ﬂoat>coherence::cohﬂt_t::staticgain [private]

The documentation for this class was generated from the following ﬁle:

•coherence.cpp

5.54 coherence::vars_t Class Reference

Inheritance diagram for coherence::vars_t:

coherence::vars_t

MHAOvlFilter::fftfb

_vars_t

Public Member Functions

•vars_t (MHAParser::parser_t ∗)

242 CONTENTS

Public Attributes

•MHAParser::kw_t tau_unit

•MHAParser::vﬂoat_t tau

•MHAParser::vﬂoat_t alpha

•MHAParser::ﬂoat_t limit

•MHAParser::vﬂoat_t mapping

•MHAParser::kw_t average

•MHAParser::bool_t invert

•MHAParser::bool_t ltgcomp

•MHAParser::vﬂoat_t ltgtau

•MHAParser::vﬂoat_t staticgain

•MHAParser::int_t delay

5.54.1 Constructor & Destructor Documentation

5.54.1.1 coherence::vars_t::vars_t (

MHAParser::parser_t ∗p)

5.54.2 Member Data Documentation

5.54.2.1 MHAParser::kw_t coherence::vars_t::tau_unit

5.54.2.2 MHAParser::vﬂoat_t coherence::vars_t::tau

5.54.2.3 MHAParser::vﬂoat_t coherence::vars_t::alpha

5.54.2.4 MHAParser::ﬂoat_t coherence::vars_t::limit

5.54.2.5 MHAParser::vﬂoat_t coherence::vars_t::mapping

5.54.2.6 MHAParser::kw_t coherence::vars_t::average

5.54.2.7 MHAParser::bool_t coherence::vars_t::invert

5.54.2.8 MHAParser::bool_t coherence::vars_t::ltgcomp

5.54.2.9 MHAParser::vﬂoat_t coherence::vars_t::ltgtau

5.54.2.10 MHAParser::vﬂoat_t coherence::vars_t::staticgain

5.54.2.11 MHAParser::int_t coherence::vars_t::delay

The documentation for this class was generated from the following ﬁle:

•coherence.cpp

5.55 combc_if_t Class Reference 243

5.55 combc_if_t Class Reference

Inheritance diagram for combc_if_t:

combc_if_t

MHAPlugin::plugin_t

< combc_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< combc_t >

Public Member Functions

•combc_if_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare (mhaconﬁg_t &)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

Private Attributes

•MHAParser::int_t outchannels

•MHAParser::bool_t interleaved

•MHAParser::string_t channel_gain_name

•MHAParser::string_t element_gain_name

244 CONTENTS

Additional Inherited Members

5.55.1 Constructor & Destructor Documentation

5.55.1.1 combc_if_t::combc_if_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.55.2 Member Function Documentation

5.55.2.1 void combc_if_t::prepare (

mhaconﬁg_t &chcfg )[virtual]

Implements MHAPlugin::plugin_t<combc_t >(p. 730).

5.55.2.2 mha_wave_t ∗combc_if_t::process (

mha_wave_t ∗s)

5.55.2.3 mha_spec_t ∗combc_if_t::process (

mha_spec_t ∗s)

5.55.3 Member Data Documentation

5.55.3.1 MHAParser::int_t combc_if_t::outchannels [private]

5.55.3.2 MHAParser::bool_t combc_if_t::interleaved [private]

5.55.3.3 MHAParser::string_t combc_if_t::channel_gain_name [private]

5.55.3.4 MHAParser::string_t combc_if_t::element_gain_name [private]

The documentation for this class was generated from the following ﬁle:

•combinechannels.cpp

5.56 combc_t Class Reference

Public Member Functions

•combc_t (algo_comm_t ac, mhaconﬁg_t cfg_input, mhaconﬁg_t cfg_output, std←-

::vector<ﬂoat >channel_gains, const std::string &element_gain_name, bool interleaved)

•mha_wave_t ∗process (mha_wave_t ∗s)

•mha_spec_t ∗process (mha_spec_t ∗s)

5.56 combc_t Class Reference 245

Private Attributes

•algo_comm_t ac_

• bool interleaved_

• unsigned int nbands

•MHASignal::waveform_t w_out

•MHASignal::spectrum_t s_out

• std::vector<mha_real_t >channel_gains_

• std::string element_gain_name_

5.56.1 Constructor & Destructor Documentation

5.56.1.1 combc_t::combc_t (

algo_comm_t ac,

mhaconﬁg_t cfg_input,

mhaconﬁg_t cfg_output,

std::vector<ﬂoat >channel_gains,

const std::string & element_gain_name,

bool interleaved )

5.56.2 Member Function Documentation

5.56.2.1 mha_wave_t ∗combc_t::process (

mha_wave_t ∗s)

5.56.2.2 mha_spec_t ∗combc_t::process (

mha_spec_t ∗s)

5.56.3 Member Data Documentation

5.56.3.1 algo_comm_t combc_t::ac_ [private]

5.56.3.2 bool combc_t::interleaved_ [private]

5.56.3.3 unsigned int combc_t::nbands [private]

5.56.3.4 MHASignal::waveform_t combc_t::w_out [private]

5.56.3.5 MHASignal::spectrum_t combc_t::s_out [private]

5.56.3.6 std::vector<mha_real_t>combc_t::channel_gains_ [private]

5.56.3.7 std::string combc_t::element_gain_name_ [private]

The documentation for this class was generated from the following ﬁle:

•combinechannels.cpp

246 CONTENTS

5.57 comm_var_t Struct Reference

Algorithm communication variable structure.

Public Attributes

• unsigned int data_type

Type of data.

• unsigned int num_entries

Number of entries.

• unsigned int stride

length of one row (C interpretation) or of one column (Fortran interpretation)

• void ∗data

Pointer to variable data.

5.57.1 Detailed Description

Algorithm communication variable structure.

Algorithm communication variables (AC variables) are objects of this type. The member data

is a pointer to the variable ‘data'. This pointer has to be valid for the lifetime of this AC variable.

The member ‘data_type' can be one of the predeﬁned types or any user deﬁned type. The

member ‘num_entries' describes the number of elements of this base type stored at the pointer

address.

An AC variable can be registered with the \ref

algo_comm_t::insert_var "insert_var" function.

5.57.2 Member Data Documentation

5.57.2.1 comm_var_t::data_type

Type of data.

This can be one of the predeﬁned types

• MHA_AC_CHAR

• MHA_AC_INT

• MHA_AC_MHAREAL

• MHA_AC_FLOAT

• MHA_AC_DOUBLE

• MHA_AC_MHACOMPLEX

• MHA_AC_VEC_FLOAT or any user deﬁned type with a value greater than

• MHA_AC_USER

5.58 cpuload_t Class Reference 247

5.57.2.2 comm_var_t::num_entries

Number of entries.

5.57.2.3 comm_var_t::stride

length of one row (C interpretation) or of one column (Fortran interpretation)

5.57.2.4 comm_var_t::data

Pointer to variable data.

The documentation for this struct was generated from the following ﬁles:

•mha.hh

•mha_algo_comm.cpp

5.58 cpuload_t Class Reference

Inheritance diagram for cpuload_t:

cpuload_t

MHAPlugin::plugin_t

< float >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< float >

248 CONTENTS

Public Member Functions

•cpuload_t (algo_comm_t, const char ∗, const char ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

Private Member Functions

• void compute_something ()

• void compute_something_else ()

Private Attributes

•MHAParser::ﬂoat_t factor

•MHAParser::bool_t use_sine

• ﬂoat phase

• volatile ﬂoat result

• std::vector<ﬂoat >table

Additional Inherited Members

5.58.1 Constructor & Destructor Documentation

5.58.1.1 cpuload_t::cpuload_t (

algo_comm_t iac,

const char ∗,

const char ∗)

5.58.2 Member Function Documentation

5.58.2.1 mha_spec_t ∗cpuload_t::process (

mha_spec_t ∗s)

5.58.2.2 mha_wave_t ∗cpuload_t::process (

mha_wave_t ∗s)

5.58.2.3 void cpuload_t::prepare (

mhaconﬁg_t &cf )[virtual]

Implements MHAPlugin::plugin_t<ﬂoat >(p. 730).

5.59 db_if_t Class Reference 249

5.58.2.4 void cpuload_t::compute_something ( ) [inline],[private]

5.58.2.5 void cpuload_t::compute_something_else ( ) [inline],[private]

5.58.3 Member Data Documentation

5.58.3.1 MHAParser::ﬂoat_t cpuload_t::factor [private]

5.58.3.2 MHAParser::bool_t cpuload_t::use_sine [private]

5.58.3.3 ﬂoat cpuload_t::phase [private]

5.58.3.4 volatile ﬂoat cpuload_t::result [private]

5.58.3.5 std::vector<ﬂoat>cpuload_t::table [private]

The documentation for this class was generated from the following ﬁle:

•cpuload.cpp

5.59 db_if_t Class Reference

Inheritance diagram for db_if_t:

db_if_t

MHAPlugin::plugin_t

< db_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< db_t >

250 CONTENTS

Public Member Functions

•db_if_t (algo_comm_t, std::string, std::string)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

•∼db_if_t ()

Private Attributes

•MHAEvents::patchbay_t<db_if_t >patchbay

•MHAParser::int_t fragsize

•MHAParser::mhapluginloader_t plugloader

• std::string chain

• std::string algo

• bool bypass

Additional Inherited Members

5.59.1 Constructor & Destructor Documentation

5.59.1.1 db_if_t::db_if_t (

algo_comm_t iac,

std::string th,

std::string al )

5.59.1.2 db_if_t::∼db_if_t ( )

5.59.2 Member Function Documentation

5.59.2.1 mha_wave_t ∗db_if_t::process (

mha_wave_t ∗s)

5.59.2.2 void db_if_t::prepare (

mhaconﬁg_t &conf )[virtual]

Implements MHAPlugin::plugin_t<db_t >(p. 730).

5.59.2.3 void db_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<db_t >(p. 731).

5.60 db_t Class Reference 251

5.59.3 Member Data Documentation

5.59.3.1 MHAEvents::patchbay_t<db_if_t >db_if_t::patchbay [private]

5.59.3.2 MHAParser::int_t db_if_t::fragsize [private]

5.59.3.3 MHAParser::mhapluginloader_t db_if_t::plugloader [private]

5.59.3.4 std::string db_if_t::chain [private]

5.59.3.5 std::string db_if_t::algo [private]

5.59.3.6 bool db_if_t::bypass [private]

The documentation for this class was generated from the following ﬁle:

•db.cpp

5.60 db_t Class Reference

Inheritance diagram for db_t:

db_t

MHASignal::doublebuffer_t

Public Member Functions

•db_t (unsigned int outer_fragsize, unsigned int inner_fragsize, unsigned int nch_in, un-

signed int nch_out, MHAParser::mhapluginloader_t &plug)

•mha_wave_t ∗inner_process (mha_wave_t ∗)

Private Attributes

•MHAParser::mhapluginloader_t &plugloader

252 CONTENTS

Additional Inherited Members

5.60.1 Constructor & Destructor Documentation

5.60.1.1 db_t::db_t (

unsigned int outer_fragsize,

unsigned int inner_fragsize,

unsigned int nch_in,

unsigned int nch_out,

MHAParser::mhapluginloader_t &plug )

5.60.2 Member Function Documentation

5.60.2.1 mha_wave_t ∗db_t::inner_process (

mha_wave_t ∗s)[virtual]

Implements MHASignal::doublebuffer_t (p. 771).

5.60.3 Member Data Documentation

5.60.3.1 MHAParser::mhapluginloader_t& db_t::plugloader [private]

The documentation for this class was generated from the following ﬁle:

•db.cpp

5.61 dc::dc_if_t Class Reference 253

5.61 dc::dc_if_t Class Reference

Inheritance diagram for dc::dc_if_t:

dc::dc_if_t

MHAPlugin::plugin_t

< dc_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< dc_t >

dc::dc_vars_t

Public Member Functions

•dc_if_t (const algo_comm_t &ac_, const std::string &th_, const std::string &al_)

• void prepare (mhaconﬁg_t &tf)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

Private Member Functions

• void update_monitors ()

Called from within the processing routines: updates the monitor variables.

• void update ()

Called by MHA conﬁguration change event mechanism: creates new runtime conﬁguration.

Private Attributes

• std::string algo

•wideband_inhib_vars_t wbinhib

•MHAEvents::patchbay_t<dc_if_t >patchbay

254 CONTENTS

Additional Inherited Members

5.61.1 Constructor & Destructor Documentation

5.61.1.1 dc::dc_if_t::dc_if_t (

const algo_comm_t &ac_,

const std::string & th_,

const std::string & al_ )

5.61.2 Member Function Documentation

5.61.2.1 void dc::dc_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<dc_t >(p. 730).

5.61.2.2 mha_wave_t ∗dc::dc_if_t::process (

mha_wave_t ∗s_in )

5.61.2.3 mha_spec_t ∗dc::dc_if_t::process (

mha_spec_t ∗s_in )

5.61.2.4 void dc::dc_if_t::update_monitors ( ) [private]

Called from within the processing routines: updates the monitor variables.

5.61.2.5 void dc::dc_if_t::update ( ) [private]

Called by MHA conﬁguration change event mechanism: creates new runtime conﬁguration.

5.61.3 Member Data Documentation

5.61.3.1 std::string dc::dc_if_t::algo [private]

5.61.3.2 wideband_inhib_vars_t dc::dc_if_t::wbinhib [private]

5.61.3.3 MHAEvents::patchbay_t<dc_if_t>dc::dc_if_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•dc.cpp

5.62 dc::dc_t Class Reference 255

5.62 dc::dc_t Class Reference

Inheritance diagram for dc::dc_t:

dc::dc_t

dc::dc_vars_validator_t

Public Member Functions

•dc_t (dc_vars_t vars, mha_real_t ﬁlter_rate, unsigned int nch, algo_comm_t ac, mha←-

_domain_t domain, unsigned int fftlen, std::string algo)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗,wb_inhib_cfg_t ∗wbinhib)

• void explicit_insert ()

• unsigned get_nbands () const

Number of frequency bands accessor.

• const MHASignal::waveform_t &get_level_in_db () const

• const MHASignal::waveform_t &get_level_in_db_adjusted () const

Private Attributes

• std::vector<MHATableLookup::linear_table_t >gt

•MHAFilter::o1ﬂt_lowpass_t rmslevel

•MHAFilter::o1ﬂt_lowpass_t attack

•MHAFilter::o1ﬂt_maxtrack_t decay

• bool powersum

• bool bypass

• unsigned int naudiochannels

• unsigned int nbands

•MHA_AC::waveform_t level_in_db

•MHA_AC::waveform_t level_in_db_adjusted

•MHA_AC::waveform_t inhib_gain

•MHASignal::waveform_t max_level_difference

• unsigned int k_nyquist

256 CONTENTS

Additional Inherited Members

5.62.1 Constructor & Destructor Documentation

5.62.1.1 dc::dc_t::dc_t (

dc_vars_t vars,

mha_real_t ﬁlter_rate,

unsigned int nch,

algo_comm_t ac,

mha_domain_t domain,

unsigned int fftlen,

std::string algo )

5.62.2 Member Function Documentation

5.62.2.1 mha_wave_t ∗dc::dc_t::process (

mha_wave_t ∗s)

5.62.2.2 mha_spec_t ∗dc::dc_t::process (

mha_spec_t ∗s,

wb_inhib_cfg_t ∗wbinhib )

5.62.2.3 void dc::dc_t::explicit_insert ( )

5.62.2.4 unsigned dc::dc_t::get_nbands ( ) const [inline]

Number of frequency bands accessor.

5.62.2.5 const MHASignal::waveform_t& dc::dc_t::get_level_in_db ( ) const [inline]

5.62.2.6 const MHASignal::waveform_t& dc::dc_t::get_level_in_db_adjusted ( ) const

[inline]

5.62.3 Member Data Documentation

5.62.3.1 std::vector<MHATableLookup::linear_table_t>dc::dc_t::gt [private]

5.62.3.2 MHAFilter::o1ﬂt_lowpass_t dc::dc_t::rmslevel [private]

5.62.3.3 MHAFilter::o1ﬂt_lowpass_t dc::dc_t::attack [private]

5.62.3.4 MHAFilter::o1ﬂt_maxtrack_t dc::dc_t::decay [private]

5.62.3.5 bool dc::dc_t::powersum [private]

5.63 dc::dc_vars_t Class Reference 257

5.62.3.6 bool dc::dc_t::bypass [private]

5.62.3.7 unsigned int dc::dc_t::naudiochannels [private]

5.62.3.8 unsigned int dc::dc_t::nbands [private]

5.62.3.9 MHA_AC::waveform_t dc::dc_t::level_in_db [private]

5.62.3.10 MHA_AC::waveform_t dc::dc_t::level_in_db_adjusted [private]

5.62.3.11 MHA_AC::waveform_t dc::dc_t::inhib_gain [private]

5.62.3.12 MHASignal::waveform_t dc::dc_t::max_level_difference [private]

5.62.3.13 unsigned int dc::dc_t::k_nyquist [private]

The documentation for this class was generated from the following ﬁle:

•dc.cpp

5.63 dc::dc_vars_t Class Reference

Inheritance diagram for dc::dc_vars_t:

dc::dc_vars_t

dc::dc_if_t

Public Member Functions

•dc_vars_t (MHAParser::parser_t &)

258 CONTENTS

Public Attributes

•MHAParser::bool_t powersum

•MHAParser::mﬂoat_t gtdata

•MHAParser::vﬂoat_t gtmin

•MHAParser::vﬂoat_t gtstep

•MHAParser::vﬂoat_t taurmslevel

•MHAParser::vﬂoat_t tauattack

•MHAParser::vﬂoat_t taudecay

•MHAParser::string_t ﬁlterbank

• std::string cf_name

• std::string ef_name

• std::string bw_name

•MHAParser::string_t chname

•MHAParser::bool_t bypass

•MHAParser::string_t clientid

•MHAParser::string_t gainrule

•MHAParser::string_t preset

•MHAParser::int_mon_t modiﬁed

•MHAParser::mﬂoat_t max_level_difference

•MHAParser::vﬂoat_mon_t input_level

•MHAParser::vﬂoat_mon_t ﬁltered_level

•MHAParser::vﬂoat_mon_t center_frequencies

•MHAParser::vﬂoat_mon_t edge_frequencies

•MHAParser::vﬂoat_mon_t band_weights

•MHAParser::bool_t use_wbinhib

5.63.1 Constructor & Destructor Documentation

5.63.1.1 dc::dc_vars_t::dc_vars_t (

MHAParser::parser_t &p)

5.63.2 Member Data Documentation

5.63.2.1 MHAParser::bool_t dc::dc_vars_t::powersum

5.63.2.2 MHAParser::mﬂoat_t dc::dc_vars_t::gtdata

5.63.2.3 MHAParser::vﬂoat_t dc::dc_vars_t::gtmin

5.63.2.4 MHAParser::vﬂoat_t dc::dc_vars_t::gtstep

5.63.2.5 MHAParser::vﬂoat_t dc::dc_vars_t::taurmslevel

5.63 dc::dc_vars_t Class Reference 259

5.63.2.6 MHAParser::vﬂoat_t dc::dc_vars_t::tauattack

5.63.2.7 MHAParser::vﬂoat_t dc::dc_vars_t::taudecay

5.63.2.8 MHAParser::string_t dc::dc_vars_t::ﬁlterbank

5.63.2.9 std::string dc::dc_vars_t::cf_name

5.63.2.10 std::string dc::dc_vars_t::ef_name

5.63.2.11 std::string dc::dc_vars_t::bw_name

5.63.2.12 MHAParser::string_t dc::dc_vars_t::chname

5.63.2.13 MHAParser::bool_t dc::dc_vars_t::bypass

5.63.2.14 MHAParser::string_t dc::dc_vars_t::clientid

5.63.2.15 MHAParser::string_t dc::dc_vars_t::gainrule

5.63.2.16 MHAParser::string_t dc::dc_vars_t::preset

5.63.2.17 MHAParser::int_mon_t dc::dc_vars_t::modiﬁed

5.63.2.18 MHAParser::mﬂoat_t dc::dc_vars_t::max_level_difference

5.63.2.19 MHAParser::vﬂoat_mon_t dc::dc_vars_t::input_level

5.63.2.20 MHAParser::vﬂoat_mon_t dc::dc_vars_t::ﬁltered_level

5.63.2.21 MHAParser::vﬂoat_mon_t dc::dc_vars_t::center_frequencies

5.63.2.22 MHAParser::vﬂoat_mon_t dc::dc_vars_t::edge_frequencies

5.63.2.23 MHAParser::vﬂoat_mon_t dc::dc_vars_t::band_weights

5.63.2.24 MHAParser::bool_t dc::dc_vars_t::use_wbinhib

The documentation for this class was generated from the following ﬁle:

•dc.cpp

260 CONTENTS

5.64 dc::dc_vars_validator_t Class Reference

Inheritance diagram for dc::dc_vars_validator_t:

dc::dc_vars_validator_t

dc::dc_t

Public Member Functions

•dc_vars_validator_t (dc_vars_t &v, unsigned int s, mha_domain_t domain)

5.64.1 Constructor & Destructor Documentation

5.64.1.1 dc::dc_vars_validator_t::dc_vars_validator_t (

dc_vars_t &v,

unsigned int s,

mha_domain_t domain )

The documentation for this class was generated from the following ﬁle:

•dc.cpp

5.65 dc::wb_inhib_cfg_t Class Reference

Public Member Functions

•wb_inhib_cfg_t (const wideband_inhib_vars_t &vars)

Public Attributes

• std::vector<ﬂoat >weights

• ﬂoat dl_map_min

• ﬂoat dl_map_max

• ﬂoat dl_diff

• ﬂoat l_min

• std::vector<std::vector<ﬂoat > > g_scale

5.66 dc::wideband_inhib_vars_t Class Reference 261

5.65.1 Constructor & Destructor Documentation

5.65.1.1 dc::wb_inhib_cfg_t::wb_inhib_cfg_t (

const wideband_inhib_vars_t &vars )

5.65.2 Member Data Documentation

5.65.2.1 std::vector<ﬂoat>dc::wb_inhib_cfg_t::weights

5.65.2.2 ﬂoat dc::wb_inhib_cfg_t::dl_map_min

5.65.2.3 ﬂoat dc::wb_inhib_cfg_t::dl_map_max

5.65.2.4 ﬂoat dc::wb_inhib_cfg_t::dl_diff

5.65.2.5 ﬂoat dc::wb_inhib_cfg_t::l_min

5.65.2.6 std::vector<std::vector<ﬂoat> > dc::wb_inhib_cfg_t::g_scale

The documentation for this class was generated from the following ﬁle:

•dc.cpp

5.66 dc::wideband_inhib_vars_t Class Reference

Inheritance diagram for dc::wideband_inhib_vars_t:

dc::wideband_inhib

_vars_t

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< wb_inhib_cfg_t >

262 CONTENTS

Public Member Functions

•wideband_inhib_vars_t ()

• void setchannels (unsigned int ch, unsigned int bnds)

•wb_inhib_cfg_t ∗current ()

• void update ()

Public Attributes

•MHAParser::vﬂoat_t weights

•MHAParser::ﬂoat_t dl_map_min

•MHAParser::ﬂoat_t dl_map_max

•MHAParser::ﬂoat_t l_min

•MHAParser::mﬂoat_t g_scale

•MHAEvents::patchbay_t<wideband_inhib_vars_t >patchbay

• unsigned int channels

• unsigned int bands

Additional Inherited Members

5.66.1 Constructor & Destructor Documentation

5.66.1.1 dc::wideband_inhib_vars_t::wideband_inhib_vars_t ( )

5.66.2 Member Function Documentation

5.66.2.1 void dc::wideband_inhib_vars_t::setchannels (

unsigned int ch,

unsigned int bnds )[inline]

5.66.2.2 wb_inhib_cfg_t∗dc::wideband_inhib_vars_t::current ( ) [inline]

5.66.2.3 void dc::wideband_inhib_vars_t::update ( )

5.66.3 Member Data Documentation

5.66.3.1 MHAParser::vﬂoat_t dc::wideband_inhib_vars_t::weights

5.66.3.2 MHAParser::ﬂoat_t dc::wideband_inhib_vars_t::dl_map_min

5.66.3.3 MHAParser::ﬂoat_t dc::wideband_inhib_vars_t::dl_map_max

5.66.3.4 MHAParser::ﬂoat_t dc::wideband_inhib_vars_t::l_min

5.67 dc_simple::dc_if_t Class Reference 263

5.66.3.5 MHAParser::mﬂoat_t dc::wideband_inhib_vars_t::g_scale

5.66.3.6 MHAEvents::patchbay_t<wideband_inhib_vars_t>dc::wideband_inhib_vars_t←-

::patchbay

5.66.3.7 unsigned int dc::wideband_inhib_vars_t::channels

5.66.3.8 unsigned int dc::wideband_inhib_vars_t::bands

The documentation for this class was generated from the following ﬁle:

•dc.cpp

5.67 dc_simple::dc_if_t Class Reference

Inheritance diagram for dc_simple::dc_if_t:

dc_simple::dc_if_t

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< runtime_cfg_t >

LEVEL dc_simple::dc_vars_t

Public Member Functions

•dc_if_t (const algo_comm_t &ac_, const std::string &th_, const std::string &al_)

• void prepare (mhaconﬁg_t &tf)

• void release ()

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

264 CONTENTS

Private Member Functions

• void update_dc ()

• void update_level ()

• void has_been_modiﬁed ()

• void read_modiﬁed ()

• void update_level_mon ()

• void update_gain_mon ()

Private Attributes

•MHAParser::string_t clientid

•MHAParser::string_t gainrule

•MHAParser::string_t preset

•MHAParser::int_mon_t modiﬁed

•MHAParser::vﬂoat_mon_t mon_l

•MHAParser::vﬂoat_mon_t mon_g

•MHAParser::string_t ﬁlterbank

•MHAParser::vﬂoat_mon_t center_frequencies

•MHAParser::vﬂoat_mon_t edge_frequencies

•MHAEvents::patchbay_t<dc_if_t >patchbay

• bool prepared

Additional Inherited Members

5.67.1 Constructor & Destructor Documentation

5.67.1.1 dc_simple::dc_if_t::dc_if_t (

const algo_comm_t &ac_,

const std::string & th_,

const std::string & al_ )

5.67.2 Member Function Documentation

5.67.2.1 void dc_simple::dc_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<runtime_cfg_t >(p. 730).

5.67.2.2 void dc_simple::dc_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<runtime_cfg_t >(p. 731).

5.67 dc_simple::dc_if_t Class Reference 265

5.67.2.3 mha_spec_t ∗dc_simple::dc_if_t::process (

mha_spec_t ∗s)

5.67.2.4 mha_wave_t ∗dc_simple::dc_if_t::process (

mha_wave_t ∗s)

5.67.2.5 void dc_simple::dc_if_t::update_dc ( ) [private]

5.67.2.6 void dc_simple::dc_if_t::update_level ( ) [private]

5.67.2.7 void dc_simple::dc_if_t::has_been_modiﬁed ( ) [inline],[private]

5.67.2.8 void dc_simple::dc_if_t::read_modiﬁed ( ) [inline],[private]

5.67.2.9 void dc_simple::dc_if_t::update_level_mon ( ) [private]

5.67.2.10 void dc_simple::dc_if_t::update_gain_mon ( ) [private]

5.67.3 Member Data Documentation

5.67.3.1 MHAParser::string_t dc_simple::dc_if_t::clientid [private]

5.67.3.2 MHAParser::string_t dc_simple::dc_if_t::gainrule [private]

5.67.3.3 MHAParser::string_t dc_simple::dc_if_t::preset [private]

5.67.3.4 MHAParser::int_mon_t dc_simple::dc_if_t::modiﬁed [private]

5.67.3.5 MHAParser::vﬂoat_mon_t dc_simple::dc_if_t::mon_l [private]

5.67.3.6 MHAParser::vﬂoat_mon_t dc_simple::dc_if_t::mon_g [private]

5.67.3.7 MHAParser::string_t dc_simple::dc_if_t::ﬁlterbank [private]

5.67.3.8 MHAParser::vﬂoat_mon_t dc_simple::dc_if_t::center_frequencies [private]

5.67.3.9 MHAParser::vﬂoat_mon_t dc_simple::dc_if_t::edge_frequencies [private]

5.67.3.10 MHAEvents::patchbay_t<dc_if_t>dc_simple::dc_if_t::patchbay [private]

5.67.3.11 bool dc_simple::dc_if_t::prepared [private]

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

266 CONTENTS

5.68 dc_simple::dc_t Class Reference

Inheritance diagram for dc_simple::dc_t:

dc_simple::dc_t

dc_simple::dc_vars

_validator_t

Classes

• class line_t

Public Member Functions

•dc_t (const dc_vars_t &vars, mha_real_t ﬁlter_rate, unsigned int nch, unsigned int fftlen)

•mha_spec_t ∗process (mha_spec_t ∗,mha_wave_t ∗level_db)

•mha_wave_t ∗process (mha_wave_t ∗,mha_wave_t ∗level_db)

Public Attributes

• std::vector<ﬂoat >mon_l

• std::vector<ﬂoat >mon_g

Private Attributes

• std::vector<mha_real_t >expansion_threshold

• std::vector<mha_real_t >limiter_threshold

• std::vector<line_t >compression

• std::vector<line_t >expansion

• std::vector<line_t >limiter

• std::vector<mha_real_t >maxgain

• unsigned int nbands

5.68 dc_simple::dc_t Class Reference 267

Additional Inherited Members

5.68.1 Constructor & Destructor Documentation

5.68.1.1 dc_simple::dc_t::dc_t (

const dc_vars_t &vars,

mha_real_t ﬁlter_rate,

unsigned int nch,

unsigned int fftlen )

5.68.2 Member Function Documentation

5.68.2.1 mha_spec_t ∗dc_simple::dc_t::process (

mha_spec_t ∗s,

mha_wave_t ∗level_db )

5.68.2.2 mha_wave_t ∗dc_simple::dc_t::process (

mha_wave_t ∗s,

mha_wave_t ∗level_db )

5.68.3 Member Data Documentation

5.68.3.1 std::vector<mha_real_t>dc_simple::dc_t::expansion_threshold [private]

5.68.3.2 std::vector<mha_real_t>dc_simple::dc_t::limiter_threshold [private]

5.68.3.3 std::vector<line_t>dc_simple::dc_t::compression [private]

5.68.3.4 std::vector<line_t>dc_simple::dc_t::expansion [private]

5.68.3.5 std::vector<line_t>dc_simple::dc_t::limiter [private]

5.68.3.6 std::vector<mha_real_t>dc_simple::dc_t::maxgain [private]

5.68.3.7 unsigned int dc_simple::dc_t::nbands [private]

5.68.3.8 std::vector<ﬂoat>dc_simple::dc_t::mon_l

5.68.3.9 std::vector<ﬂoat>dc_simple::dc_t::mon_g

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

268 CONTENTS

5.69 dc_simple::dc_t::line_t Class Reference

Public Member Functions

•line_t (mha_real_t x1, mha_real_t y1, mha_real_t x2, mha_real_t y2)

•line_t (mha_real_t x1, mha_real_t y1, mha_real_t slope)

•mha_real_t operator() (mha_real_t x)

Private Attributes

•mha_real_t m

•mha_real_t y0

5.69.1 Constructor & Destructor Documentation

5.69.1.1 dc_simple::dc_t::line_t::line_t (

mha_real_t x1,

mha_real_t y1,

mha_real_t x2,

mha_real_t y2 )

5.69.1.2 dc_simple::dc_t::line_t::line_t (

mha_real_t x1,

mha_real_t y1,

mha_real_t slope )

5.69.2 Member Function Documentation

5.69.2.1 mha_real_t dc_simple::dc_t::line_t::operator() (

mha_real_t x)[inline]

5.69.3 Member Data Documentation

5.69.3.1 mha_real_t dc_simple::dc_t::line_t::m [private]

5.69.3.2 mha_real_t dc_simple::dc_t::line_t::y0 [private]

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

5.70 dc_simple::dc_vars_t Class Reference 269

5.70 dc_simple::dc_vars_t Class Reference

Inheritance diagram for dc_simple::dc_vars_t:

dc_simple::dc_vars_t

dc_simple::dc_if_t

Public Member Functions

•dc_vars_t (MHAParser::parser_t &)

Public Attributes

•MHAParser::vﬂoat_t g50

•MHAParser::vﬂoat_t g80

•MHAParser::vﬂoat_t maxgain

•MHAParser::vﬂoat_t expansion_threshold

•MHAParser::vﬂoat_t expansion_slope

•MHAParser::vﬂoat_t limiter_threshold

•MHAParser::vﬂoat_t tauattack

•MHAParser::vﬂoat_t taudecay

•MHAParser::bool_t bypass

5.70.1 Constructor & Destructor Documentation

5.70.1.1 dc_simple::dc_vars_t::dc_vars_t (

MHAParser::parser_t &p)

5.70.2 Member Data Documentation

5.70.2.1 MHAParser::vﬂoat_t dc_simple::dc_vars_t::g50

5.70.2.2 MHAParser::vﬂoat_t dc_simple::dc_vars_t::g80

270 CONTENTS

5.70.2.3 MHAParser::vﬂoat_t dc_simple::dc_vars_t::maxgain

5.70.2.4 MHAParser::vﬂoat_t dc_simple::dc_vars_t::expansion_threshold

5.70.2.5 MHAParser::vﬂoat_t dc_simple::dc_vars_t::expansion_slope

5.70.2.6 MHAParser::vﬂoat_t dc_simple::dc_vars_t::limiter_threshold

5.70.2.7 MHAParser::vﬂoat_t dc_simple::dc_vars_t::tauattack

5.70.2.8 MHAParser::vﬂoat_t dc_simple::dc_vars_t::taudecay

5.70.2.9 MHAParser::bool_t dc_simple::dc_vars_t::bypass

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

5.71 dc_simple::dc_vars_validator_t Class Reference

Inheritance diagram for dc_simple::dc_vars_validator_t:

dc_simple::dc_vars

_validator_t

dc_simple::dc_t dc_simple::level_smoother_t

Public Member Functions

•dc_vars_validator_t (const dc_vars_t &v, unsigned int s)

5.72 dc_simple::level_smoother_t Class Reference 271

5.71.1 Constructor & Destructor Documentation

5.71.1.1 dc_simple::dc_vars_validator_t::dc_vars_validator_t (

const dc_vars_t &v,

unsigned int s)

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

5.72 dc_simple::level_smoother_t Class Reference

Inheritance diagram for dc_simple::level_smoother_t:

dc_simple::level_smoother_t

dc_simple::dc_vars

_validator_t

Public Member Functions

•level_smoother_t (const dc_vars_t &vars, mha_real_t ﬁlter_rate, mhaconﬁg_t buscfg)

•mha_wave_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

Private Attributes

•MHAFilter::o1ﬂt_lowpass_t attack

•MHAFilter::o1ﬂt_maxtrack_t decay

• unsigned int nbands

• unsigned int fftlen

•MHASignal::waveform_t level_wave

•MHASignal::waveform_t level_spec

272 CONTENTS

Additional Inherited Members

5.72.1 Constructor & Destructor Documentation

5.72.1.1 dc_simple::level_smoother_t::level_smoother_t (

const dc_vars_t &vars,

mha_real_t ﬁlter_rate,

mhaconﬁg_t buscfg )

5.72.2 Member Function Documentation

5.72.2.1 mha_wave_t ∗dc_simple::level_smoother_t::process (

mha_spec_t ∗s)

5.72.2.2 mha_wave_t ∗dc_simple::level_smoother_t::process (

mha_wave_t ∗s)

5.72.3 Member Data Documentation

5.72.3.1 MHAFilter::o1ﬂt_lowpass_t dc_simple::level_smoother_t::attack [private]

5.72.3.2 MHAFilter::o1ﬂt_maxtrack_t dc_simple::level_smoother_t::decay [private]

5.72.3.3 unsigned int dc_simple::level_smoother_t::nbands [private]

5.72.3.4 unsigned int dc_simple::level_smoother_t::fftlen [private]

5.72.3.5 MHASignal::waveform_t dc_simple::level_smoother_t::level_wave [private]

5.72.3.6 MHASignal::waveform_t dc_simple::level_smoother_t::level_spec [private]

The documentation for this class was generated from the following ﬁle:

•dc_simple.cpp

5.73 delay::interface_t Class Reference 273

5.73 delay::interface_t Class Reference

Inheritance diagram for delay::interface_t:

delay::interface_t

MHAPlugin::plugin_t

< MHASignal::delay_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< MHASignal::delay_t >

Public Member Functions

•interface_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare (mhaconﬁg_t &)

•mha_wave_t ∗process (mha_wave_t ∗)

Private Member Functions

• void update ()

Private Attributes

•MHAParser::vint_t delays

•MHAEvents::patchbay_t<interface_t >patchbay

274 CONTENTS

Additional Inherited Members

5.73.1 Constructor & Destructor Documentation

5.73.1.1 delay::interface_t::interface_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.73.2 Member Function Documentation

5.73.2.1 void delay::interface_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<MHASignal::delay_t >(p. 730).

5.73.2.2 mha_wave_t ∗delay::interface_t::process (

mha_wave_t ∗s)

5.73.2.3 void delay::interface_t::update ( ) [private]

5.73.3 Member Data Documentation

5.73.3.1 MHAParser::vint_t delay::interface_t::delays [private]

5.73.3.2 MHAEvents::patchbay_t<interface_t>delay::interface_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•delay.cpp

5.74 delaysum::delaysum_if_t Class Reference

Interface class for the delaysum plugin.

5.74 delaysum::delaysum_if_t Class Reference 275

Inheritance diagram for delaysum::delaysum_if_t:

delaysum::delaysum_if_t

MHAPlugin::plugin_t

< delaysum_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< delaysum_t >

Public Member Functions

•delaysum_if_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAParser::vﬂoat_t weights

Linear weights to be multiplied with the audio signal, one factor for each channel.

•MHAParser::vint_t delay

vector of channel-speciﬁc delays, in samples.

•MHAEvents::patchbay_t<delaysum_if_t >patchbay

The patchbay to react to conﬁg changes.

276 CONTENTS

Additional Inherited Members

5.74.1 Detailed Description

Interface class for the delaysum plugin.

This plugin allows to delay and sum multiple input channels using individual delays and weights.

After each channel gets delayed it is multiplied with the given weight and then added to the

single outout channel.

5.74.2 Constructor & Destructor Documentation

5.74.2.1 delaysum::delaysum_if_t::delaysum_if_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.74.3 Member Function Documentation

5.74.3.1 mha_wave_t ∗delaysum::delaysum_if_t::process (

mha_wave_t ∗wave )

5.74.3.2 void delaysum::delaysum_if_t::prepare (

mhaconﬁg_t &tfcfg )[virtual]

Implements MHAPlugin::plugin_t<delaysum_t >(p. 730).

5.74.3.3 void delaysum::delaysum_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<delaysum_t >(p. 731).

5.74.3.4 void delaysum::delaysum_if_t::update_cfg ( ) [private]

5.74.4 Member Data Documentation

5.74.4.1 MHAParser::vﬂoat_t delaysum::delaysum_if_t::weights [private]

Linear weights to be multiplied with the audio signal, one factor for each channel.

Order is [chan0, chan1, ...]

5.74.4.2 MHAParser::vint_t delaysum::delaysum_if_t::delay [private]

vector of channel-speciﬁc delays, in samples.

5.75 delaysum::delaysum_t Class Reference 277

5.74.4.3 MHAEvents::patchbay_t<delaysum_if_t>delaysum::delaysum_if_t::patchbay

[private]

The patchbay to react to conﬁg changes.

The documentation for this class was generated from the following ﬁle:

•delaysum.cpp

5.75 delaysum::delaysum_t Class Reference

Runtime conﬁguration of the delaysum plugin.

Inheritance diagram for delaysum::delaysum_t:

delaysum::delaysum_t

MHASignal::delay_t

Public Member Functions

•delaysum_t (unsigned int nch, unsigned int fragsize, const std::vector<mha_real_t >

&weights_, const std::vector<int >&delays_)

Constructor of the runtime conﬁguration.

•mha_wave_t ∗process (mha_wave_t ∗)

Private Attributes

• std::vector<mha_real_t >weights

Relative weights for each channel. Order is [chan0, chan1, ...].

•MHASignal::waveform_t out

Output waveform.

278 CONTENTS

5.75.1 Detailed Description

Runtime conﬁguration of the delaysum plugin.

Inherits from the already present delay_t class. The constructor initializes and validates the

runtime conﬁguration and forwards the delay vector to the delay_t class. The process function

ﬁrst calls delay_t::process and then multiplies every output channel with its weight and adds

them into the output channel.

5.75.2 Constructor & Destructor Documentation

5.75.2.1 delaysum::delaysum_t::delaysum_t (

unsigned int nch,

unsigned int fragsize,

const std::vector<mha_real_t >&weights_,

const std::vector<int >&delays_ )

Constructor of the runtime conﬁguration.

Parameters

nch Number of input channels.

fragsize Size of one input fragment in frames.

weights←-

Vector of weights for each channel.

delays←-

Vector of delays, one entry per channel.

5.75.3 Member Function Documentation

5.75.3.1 mha_wave_t ∗delaysum::delaysum_t::process (

mha_wave_t ∗signal )

5.75.4 Member Data Documentation

5.75.4.1 std::vector<mha_real_t>delaysum::delaysum_t::weights [private]

Relative weights for each channel. Order is [chan0, chan1, ...].

5.75.4.2 MHASignal::waveform_t delaysum::delaysum_t::out [private]

Output waveform.

The documentation for this class was generated from the following ﬁle:

•delaysum.cpp

5.76 doasvm_classiﬁcation Class Reference 279

5.76 doasvm_classiﬁcation Class Reference

Inheritance diagram for doasvm_classiﬁcation:

doasvm_classification

MHAPlugin::plugin_t

< doasvm_classification

_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< doasvm_classification

_config >

Public Member Functions

•doasvm_classiﬁcation (algo_comm_t &ac, const std::string &chain_name, const std←-

::string &algo_name)

Constructs our plugin.

•∼doasvm_classiﬁcation ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::vﬂoat_t angles

•MHAParser::mﬂoat_t w

•MHAParser::vﬂoat_t b

•MHAParser::vﬂoat_t x

•MHAParser::vﬂoat_t y

•MHAParser::string_t p_name

•MHAParser::string_t max_p_ind_name

•MHAParser::string_t vGCC_name

280 CONTENTS

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<doasvm_classiﬁcation >patchbay

Additional Inherited Members

5.76.1 Constructor & Destructor Documentation

5.76.1.1 doasvm_classiﬁcation::doasvm_classiﬁcation (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.76.1.2 doasvm_classiﬁcation::∼doasvm_classiﬁcation ( )

5.76.2 Member Function Documentation

5.76.2.1 mha_wave_t ∗doasvm_classiﬁcation::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.76.2.2 void doasvm_classiﬁcation::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<doasvm_classiﬁcation_conﬁg >(p. 730).

5.77 doasvm_classiﬁcation_conﬁg Class Reference 281

5.76.2.3 void doasvm_classiﬁcation::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<doasvm_classiﬁcation_conﬁg >(p. 731).

5.76.2.4 void doasvm_classiﬁcation::update_cfg ( ) [private]

5.76.3 Member Data Documentation

5.76.3.1 MHAParser::vﬂoat_t doasvm_classiﬁcation::angles

5.76.3.2 MHAParser::mﬂoat_t doasvm_classiﬁcation::w

5.76.3.3 MHAParser::vﬂoat_t doasvm_classiﬁcation::b

5.76.3.4 MHAParser::vﬂoat_t doasvm_classiﬁcation::x

5.76.3.5 MHAParser::vﬂoat_t doasvm_classiﬁcation::y

5.76.3.6 MHAParser::string_t doasvm_classiﬁcation::p_name

5.76.3.7 MHAParser::string_t doasvm_classiﬁcation::max_p_ind_name

5.76.3.8 MHAParser::string_t doasvm_classiﬁcation::vGCC_name

5.76.3.9 MHAEvents::patchbay_t<doasvm_classiﬁcation>doasvm_classiﬁcation::patchbay

[private]

The documentation for this class was generated from the following ﬁles:

•doasvm_classiﬁcation.h

•doasvm_classiﬁcation.cpp

5.77 doasvm_classiﬁcation_conﬁg Class Reference

Public Member Functions

•doasvm_classiﬁcation_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg,

doasvm_classiﬁcation ∗_doasvm)

•∼doasvm_classiﬁcation_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

282 CONTENTS

Public Attributes

•algo_comm_t &ac

•doasvm_classiﬁcation ∗doasvm

•MHA_AC::waveform_t p

•MHA_AC::int_t p_max

•mha_wave_t c

5.77.1 Constructor & Destructor Documentation

5.77.1.1 doasvm_classiﬁcation_conﬁg::doasvm_classiﬁcation_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

doasvm_classiﬁcation ∗_doasvm )

5.77.1.2 doasvm_classiﬁcation_conﬁg::∼doasvm_classiﬁcation_conﬁg ( )

5.77.2 Member Function Documentation

5.77.2.1 mha_wave_t ∗doasvm_classiﬁcation_conﬁg::process (

mha_wave_t ∗wave )

5.77.3 Member Data Documentation

5.77.3.1 algo_comm_t& doasvm_classiﬁcation_conﬁg::ac

5.77.3.2 doasvm_classiﬁcation∗doasvm_classiﬁcation_conﬁg::doasvm

5.77.3.3 MHA_AC::waveform_t doasvm_classiﬁcation_conﬁg::p

5.77.3.4 MHA_AC::int_t doasvm_classiﬁcation_conﬁg::p_max

5.77.3.5 mha_wave_t doasvm_classiﬁcation_conﬁg::c

The documentation for this class was generated from the following ﬁles:

•doasvm_classiﬁcation.h

•doasvm_classiﬁcation.cpp

5.78 doasvm_feature_extraction Class Reference 283

5.78 doasvm_feature_extraction Class Reference

Inheritance diagram for doasvm_feature_extraction:

doasvm_feature_extraction

MHAPlugin::plugin_t

< doasvm_feature_extraction

_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< doasvm_feature_extraction

_config >

Public Member Functions

•doasvm_feature_extraction (algo_comm_t &ac, const std::string &chain_name, const

std::string &algo_name)

Constructs our plugin.

•∼doasvm_feature_extraction ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::int_t fftlen

•MHAParser::int_t max_lag

•MHAParser::int_t nupsample

•MHAParser::string_t vGCC_name

284 CONTENTS

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<doasvm_feature_extraction >patchbay

Additional Inherited Members

5.78.1 Constructor & Destructor Documentation

5.78.1.1 doasvm_feature_extraction::doasvm_feature_extraction (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.78.1.2 doasvm_feature_extraction::∼doasvm_feature_extraction ( )

5.78.2 Member Function Documentation

5.78.2.1 mha_wave_t ∗doasvm_feature_extraction::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.78.2.2 void doasvm_feature_extraction::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<doasvm_feature_extraction_conﬁg >(p. 730).

5.79 doasvm_feature_extraction_conﬁg Class Reference 285

5.78.2.3 void doasvm_feature_extraction::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<doasvm_feature_extraction_conﬁg >

(p. 731).

5.78.2.4 void doasvm_feature_extraction::update_cfg ( ) [private]

5.78.3 Member Data Documentation

5.78.3.1 MHAParser::int_t doasvm_feature_extraction::fftlen

5.78.3.2 MHAParser::int_t doasvm_feature_extraction::max_lag

5.78.3.3 MHAParser::int_t doasvm_feature_extraction::nupsample

5.78.3.4 MHAParser::string_t doasvm_feature_extraction::vGCC_name

5.78.3.5 MHAEvents::patchbay_t<doasvm_feature_extraction>

doasvm_feature_extraction::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•doasvm_feature_extraction.h

•doasvm_feature_extraction.cpp

5.79 doasvm_feature_extraction_conﬁg Class Reference

Public Member Functions

•doasvm_feature_extraction_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg,

doasvm_feature_extraction ∗_doagcc)

•∼doasvm_feature_extraction_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

Public Attributes

•doasvm_feature_extraction ∗doagcc

• unsigned int wndlen

• unsigned int fftlen

• unsigned int G_length

• unsigned int GCC_start

• unsigned int GCC_end

•MHA_AC::waveform_t vGCC_ac

•mha_fft_t fft

•mha_fft_t ifft

• double hifftwin_sum

•MHASignal::waveform_t proc_wave

•MHASignal::waveform_t hwin

•MHASignal::waveform_t hifftwin

•MHASignal::waveform_t vGCC

•MHASignal::spectrum_t in_spec

•MHASignal::spectrum_t G

286 CONTENTS

5.79.1 Constructor & Destructor Documentation

5.79.1.1 doasvm_feature_extraction_conﬁg::doasvm_feature_extraction_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

doasvm_feature_extraction ∗_doagcc )

5.79.1.2 doasvm_feature_extraction_conﬁg::∼doasvm_feature_extraction_conﬁg ( )

5.79.2 Member Function Documentation

5.79.2.1 mha_wave_t ∗doasvm_feature_extraction_conﬁg::process (

mha_wave_t ∗wave )

5.79.3 Member Data Documentation

5.79.3.1 doasvm_feature_extraction∗doasvm_feature_extraction_conﬁg::doagcc

5.79.3.2 unsigned int doasvm_feature_extraction_conﬁg::wndlen

5.79.3.3 unsigned int doasvm_feature_extraction_conﬁg::fftlen

5.79.3.4 unsigned int doasvm_feature_extraction_conﬁg::G_length

5.79.3.5 unsigned int doasvm_feature_extraction_conﬁg::GCC_start

5.79.3.6 unsigned int doasvm_feature_extraction_conﬁg::GCC_end

5.79.3.7 MHA_AC::waveform_t doasvm_feature_extraction_conﬁg::vGCC_ac

5.79.3.8 mha_fft_t doasvm_feature_extraction_conﬁg::fft

5.79.3.9 mha_fft_t doasvm_feature_extraction_conﬁg::ifft

5.79.3.10 double doasvm_feature_extraction_conﬁg::hifftwin_sum

5.79.3.11 MHASignal::waveform_t doasvm_feature_extraction_conﬁg::proc_wave

5.79.3.12 MHASignal::waveform_t doasvm_feature_extraction_conﬁg::hwin

5.79.3.13 MHASignal::waveform_t doasvm_feature_extraction_conﬁg::hifftwin

5.79.3.14 MHASignal::waveform_t doasvm_feature_extraction_conﬁg::vGCC

5.79.3.15 MHASignal::spectrum_t doasvm_feature_extraction_conﬁg::in_spec

5.79.3.16 MHASignal::spectrum_t doasvm_feature_extraction_conﬁg::G

The documentation for this class was generated from the following ﬁles:

•doasvm_feature_extraction.h

•doasvm_feature_extraction.cpp

5.80 droptect_t Class Reference 287

5.80 droptect_t Class Reference

Detect dropouts in a signal with a constant spectrum.

Inheritance diagram for droptect_t:

droptect_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•droptect_t (algo_comm_t &ac, const std::string &chain_name, const std::string &algo←-

_name)

This constructor initializes the conﬁguration language variables and inserts them into the MHA

conﬁguration tree.

• void prepare (mhaconﬁg_t &signal_info)

• void release (void)

•mha_spec_t ∗process (mha_spec_t ∗signal)

Private Attributes

•MHAParser::vint_mon_t dropouts

•MHAParser::vint_mon_t consecutive_dropouts

•MHAParser::int_mon_t blocks

•MHAParser::bool_t reset

•MHAParser::ﬂoat_t threshold

•MHASignal::waveform_t ∗current_powspec

288 CONTENTS

•MHASignal::waveform_t ∗ﬁltered_powspec

•MHAParser::ﬂoat_t tau

• std::vector<bool >ﬁlter_activated

• ﬂoat period

The period of the process callback.

•MHAParser::mﬂoat_mon_t ﬁltered_powspec_mon

User access to ﬁltered spectrum.

•MHAParser::ﬂoat_mon_t level_mon

Additional Inherited Members

5.80.1 Detailed Description

Detect dropouts in a signal with a constant spectrum.

5.80.2 Constructor & Destructor Documentation

5.80.2.1 droptect_t::droptect_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

This constructor initializes the conﬁguration language variables and inserts them into the MHA

conﬁguration tree.

5.80.3 Member Function Documentation

5.80.3.1 void droptect_t::prepare (

mhaconﬁg_t &signal_info )[virtual]

Implements MHAPlugin::plugin_t<int >(p. 730).

5.80.3.2 void droptect_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.80 droptect_t Class Reference 289

5.80.3.3 mha_spec_t ∗droptect_t::process (

mha_spec_t ∗signal )

5.80.4 Member Data Documentation

5.80.4.1 MHAParser::vint_mon_t droptect_t::dropouts [private]

5.80.4.2 MHAParser::vint_mon_t droptect_t::consecutive_dropouts [private]

5.80.4.3 MHAParser::int_mon_t droptect_t::blocks [private]

5.80.4.4 MHAParser::bool_t droptect_t::reset [private]

5.80.4.5 MHAParser::ﬂoat_t droptect_t::threshold [private]

5.80.4.6 MHASignal::waveform_t∗droptect_t::current_powspec [private]

5.80.4.7 MHASignal::waveform_t∗droptect_t::ﬁltered_powspec [private]

5.80.4.8 MHAParser::ﬂoat_t droptect_t::tau [private]

5.80.4.9 std::vector<bool>droptect_t::ﬁlter_activated [private]

5.80.4.10 ﬂoat droptect_t::period [private]

The period of the process callback.

5.80.4.11 MHAParser::mﬂoat_mon_t droptect_t::ﬁltered_powspec_mon [private]

User access to ﬁltered spectrum.

5.80.4.12 MHAParser::ﬂoat_mon_t droptect_t::level_mon [private]

The documentation for this class was generated from the following ﬁle:

•droptect.cpp

290 CONTENTS

5.81 ds_t Class Reference

Inheritance diagram for ds_t:

ds_t

MHAPlugin::plugin_t

< MHASignal::waveform_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< MHASignal::waveform_t >

Public Member Functions

•ds_t (algo_comm_t, std::string, std::string)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Attributes

•MHAParser::int_t ratio

•MHAFilter::iir_ﬁlter_t antialias

Additional Inherited Members

5.81.1 Constructor & Destructor Documentation

5.82 dynamiclib_t Class Reference 291

5.81.1.1 ds_t::ds_t (

algo_comm_t iac,

std::string ,

std::string )

5.81.2 Member Function Documentation

5.81.2.1 mha_wave_t ∗ds_t::process (

mha_wave_t ∗s)

5.81.2.2 void ds_t::prepare (

mhaconﬁg_t &cf )[virtual]

Implements MHAPlugin::plugin_t<MHASignal::waveform_t >(p. 730).

5.81.2.3 void ds_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<MHASignal::waveform_t >(p. 731).

5.81.3 Member Data Documentation

5.81.3.1 MHAParser::int_t ds_t::ratio [private]

5.81.3.2 MHAFilter::iir_ﬁlter_t ds_t::antialias [private]

The documentation for this class was generated from the following ﬁle:

•downsample.cpp

5.82 dynamiclib_t Class Reference

Public Member Functions

•dynamiclib_t (const std::string &)

• void ∗resolve (const std::string &)

• void ∗resolve_checked (const std::string &)

•∼dynamiclib_t ()

• const std::string & getmodulename () const

• const std::string & getname () const

292 CONTENTS

Private Attributes

• std::string fullname

• std::string modulename

•mha_libhandle_t h

5.82.1 Constructor & Destructor Documentation

5.82.1.1 dynamiclib_t::dynamiclib_t (

const std::string & n)

5.82.1.2 dynamiclib_t::∼dynamiclib_t ( )

5.82.2 Member Function Documentation

5.82.2.1 void ∗dynamiclib_t::resolve (

const std::string & n)

5.82.2.2 void ∗dynamiclib_t::resolve_checked (

const std::string & n)

5.82.2.3 const std::string& dynamiclib_t::getmodulename ( ) const [inline]

5.82.2.4 const std::string& dynamiclib_t::getname ( ) const [inline]

5.82.3 Member Data Documentation

5.82.3.1 std::string dynamiclib_t::fullname [private]

5.82.3.2 std::string dynamiclib_t::modulename [private]

5.82.3.3 mha_libhandle_t dynamiclib_t::h [private]

The documentation for this class was generated from the following ﬁles:

•mha_os.h

•mha_os.cpp

5.83 DynComp::dc_afterburn_rt_t Class Reference

Real-time class for after burn effect.

5.83 DynComp::dc_afterburn_rt_t Class Reference 293

Public Member Functions

•dc_afterburn_rt_t (const std::vector<ﬂoat >&cf, unsigned int channels, ﬂoat srate,

const dc_afterburn_vars_t &vars)

• void burn (ﬂoat &Gin, ﬂoat Lin, unsigned int band, unsigned int channel)

gain modiﬁer method (afterburn).

Private Attributes

• std::vector<ﬂoat >drain_inv

• std::vector<ﬂoat >conﬂux

• std::vector<ﬂoat >maxgain

• std::vector<ﬂoat >mpo_inv

• std::vector<MHAFilter::o1ﬂt_lowpass_t >lp

5.83.1 Detailed Description

Real-time class for after burn effect.

The constructor processes the parameters and creates pre-processed variables for efﬁcient

realtime processing.

5.83.2 Constructor & Destructor Documentation

5.83.2.1 DynComp::dc_afterburn_rt_t::dc_afterburn_rt_t (

const std::vector<ﬂoat >&cf,

unsigned int channels,

ﬂoat srate,

const dc_afterburn_vars_t &vars )

5.83.3 Member Function Documentation

5.83.3.1 void DynComp::dc_afterburn_rt_t::burn (

ﬂoat & Gin,

ﬂoat Lin,

unsigned int band,

unsigned int channel )[inline]

gain modiﬁer method (afterburn).

Parameters

Gin Linear gain.

Lin Input level (Pascal).

band Filter band number.

channel Channel number.

294 CONTENTS

Output level for MPO is estimated by Gin ∗Lin.

5.83.4 Member Data Documentation

5.83.4.1 std::vector<ﬂoat>DynComp::dc_afterburn_rt_t::drain_inv [private]

5.83.4.2 std::vector<ﬂoat>DynComp::dc_afterburn_rt_t::conﬂux [private]

5.83.4.3 std::vector<ﬂoat>DynComp::dc_afterburn_rt_t::maxgain [private]

5.83.4.4 std::vector<ﬂoat>DynComp::dc_afterburn_rt_t::mpo_inv [private]

5.83.4.5 std::vector<MHAFilter::o1ﬂt_lowpass_t>DynComp::dc_afterburn_rt_t::lp [private]

The documentation for this class was generated from the following ﬁles:

•dc_afterburn.h

•dc_afterburn.cpp

5.84 DynComp::dc_afterburn_t Class Reference

Afterburn class, to be deﬁned as a member of compressors.

Inheritance diagram for DynComp::dc_afterburn_t:

DynComp::dc_afterburn_t

DynComp::dc_afterburn

_vars_t

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< dc_afterburn_rt_t >

5.84 DynComp::dc_afterburn_t Class Reference 295

Public Member Functions

•dc_afterburn_t ()

• void set_fb_pars (const std::vector<ﬂoat >&cf, unsigned int channels, ﬂoat srate)

• void unset_fb_pars ()

• void update_burner ()

• void burn (ﬂoat &Gin, ﬂoat Lin, unsigned int band, unsigned int channel)

Private Member Functions

• void update ()

Private Attributes

•MHAEvents::patchbay_t<dc_afterburn_t >patchbay

• std::vector<ﬂoat >_cf

• unsigned int _channels

• ﬂoat _srate

• bool commit_pending

• bool fb_pars_conﬁgured

Additional Inherited Members

5.84.1 Detailed Description

Afterburn class, to be deﬁned as a member of compressors.

5.84.2 Constructor & Destructor Documentation

5.84.2.1 DynComp::dc_afterburn_t::dc_afterburn_t ( )

5.84.3 Member Function Documentation

5.84.3.1 void DynComp::dc_afterburn_t::set_fb_pars (

const std::vector<ﬂoat >&cf,

unsigned int channels,

ﬂoat srate )

5.84.3.2 void DynComp::dc_afterburn_t::unset_fb_pars ( )

5.84.3.3 void DynComp::dc_afterburn_t::update_burner ( ) [inline]

296 CONTENTS

5.84.3.4 void DynComp::dc_afterburn_t::burn (

ﬂoat & Gin,

ﬂoat Lin,

unsigned int band,

unsigned int channel )[inline]

5.84.3.5 void DynComp::dc_afterburn_t::update ( ) [private]

5.84.4 Member Data Documentation

5.84.4.1 MHAEvents::patchbay_t<dc_afterburn_t>DynComp::dc_afterburn_t::patchbay

[private]

5.84.4.2 std::vector<ﬂoat>DynComp::dc_afterburn_t::_cf [private]

5.84.4.3 unsigned int DynComp::dc_afterburn_t::_channels [private]

5.84.4.4 ﬂoat DynComp::dc_afterburn_t::_srate [private]

5.84.4.5 bool DynComp::dc_afterburn_t::commit_pending [private]

5.84.4.6 bool DynComp::dc_afterburn_t::fb_pars_conﬁgured [private]

The documentation for this class was generated from the following ﬁles:

•dc_afterburn.h

•dc_afterburn.cpp

5.85 DynComp::dc_afterburn_vars_t Class Reference

Variables for dc_afterburn_t (p. 294) class.

5.85 DynComp::dc_afterburn_vars_t Class Reference 297

Inheritance diagram for DynComp::dc_afterburn_vars_t:

DynComp::dc_afterburn

_vars_t

DynComp::dc_afterburn_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•dc_afterburn_vars_t ()

Public Attributes

•MHAParser::vﬂoat_t f

•MHAParser::vﬂoat_t drain

•MHAParser::vﬂoat_t conﬂux

•MHAParser::vﬂoat_t maxgain

•MHAParser::vﬂoat_t mpo

•MHAParser::ﬂoat_t taugain

•MHAParser::kw_t commit

•MHAParser::bool_t bypass

Additional Inherited Members

5.85.1 Detailed Description

Variables for dc_afterburn_t (p. 294) class.

298 CONTENTS

5.85.2 Constructor & Destructor Documentation

5.85.2.1 DynComp::dc_afterburn_vars_t::dc_afterburn_vars_t ( )

5.85.3 Member Data Documentation

5.85.3.1 MHAParser::vﬂoat_t DynComp::dc_afterburn_vars_t::f

5.85.3.2 MHAParser::vﬂoat_t DynComp::dc_afterburn_vars_t::drain

5.85.3.3 MHAParser::vﬂoat_t DynComp::dc_afterburn_vars_t::conﬂux

5.85.3.4 MHAParser::vﬂoat_t DynComp::dc_afterburn_vars_t::maxgain

5.85.3.5 MHAParser::vﬂoat_t DynComp::dc_afterburn_vars_t::mpo

5.85.3.6 MHAParser::ﬂoat_t DynComp::dc_afterburn_vars_t::taugain

5.85.3.7 MHAParser::kw_t DynComp::dc_afterburn_vars_t::commit

5.85.3.8 MHAParser::bool_t DynComp::dc_afterburn_vars_t::bypass

The documentation for this class was generated from the following ﬁles:

•dc_afterburn.h

•dc_afterburn.cpp

5.86 DynComp::gaintable_t Class Reference

Gain table class.

5.86 DynComp::gaintable_t Class Reference 299

Public Member Functions

•gaintable_t (const std::vector<mha_real_t >&LInput, const std::vector<mha_real_t

>&FCenter, unsigned int channels)

Constructor.

•∼gaintable_t ()

• void update (std::vector<std::vector<std::vector<mha_real_t >>>newGain)

Update gains from an external table.

•mha_real_t get_gain (mha_real_t Lin, mha_real_t Fin, unsigned int channel)

Read Gain from gain table.

•mha_real_t get_gain (mha_real_t Lin, unsigned int band, unsigned int channel)

Read Gain from gain table.

• void get_gain (const mha_wave_t &Lin, mha_wave_t &Gain)

Read Gains from gain table.

• unsigned int nbands () const

Return number of frequency bands.

• unsigned int nchannels () const

Return number of audio channels.

• std::vector<std::vector<mha_real_t > > get_iofun () const

Return current input-output function.

• std::vector<mha_real_t >get_vL () const

• std::vector<mha_real_t >get_vF () const

Private Attributes

• unsigned int num_L

• unsigned int num_F

• unsigned int num_channels

• std::vector<mha_real_t >vL

• std::vector<mha_real_t >vF

• std::vector<mha_real_t >vFlog

• std::vector<std::vector<std::vector<mha_real_t >>>data

5.86.1 Detailed Description

Gain table class.

This gain table is intended to efﬁcient table lookup, i.e, interpolation of levels, and optional

interpolation of frequencies. Sample input levels and sample frequencies are given in the con-

structor. The gain entries can be updated with the update() (p. 301) member function via a gain

prescription rule from an auditory proﬁle.

300 CONTENTS

5.86.2 Constructor & Destructor Documentation

5.86.2.1 gaintable_t::gaintable_t (

const std::vector<mha_real_t >&LInput,

const std::vector<mha_real_t >&FCenter,

unsigned int channels )

Constructor.

5.86 DynComp::gaintable_t Class Reference 301

Parameters

LInput Input level samples, in equivalent LTASS_combined dB SPL.

FCenter Frequency samples in Hz (e.g., center frequencies of ﬁlterbank).

channels Number of audio channels (typically 2).

5.86.2.2 gaintable_t::∼gaintable_t ( )

5.86.3 Member Function Documentation

5.86.3.1 void gaintable_t::update (

std::vector<std::vector<std::vector<mha_real_t >>>newGain )

Update gains from an external table.

Parameters

newGain New gain table entries.

Dimension change is not allowed. The number of entries are checked.

5.86.3.2 mha_real_t gaintable_t::get_gain (

mha_real_t Lin,

mha_real_t Fin,

unsigned int channel )

Read Gain from gain table.

Parameters

Lin Input level

Fin Input frequency (no match required)

channel Audio channel

5.86.3.3 mha_real_t gaintable_t::get_gain (

mha_real_t Lin,

unsigned int band,

unsigned int channel )

Read Gain from gain table.

Parameters

Lin Input level

302 CONTENTS

Parameters

band Input frequency band

channel Audio channel

5.86.3.4 void gaintable_t::get_gain (

const mha_wave_t &Lin,

mha_wave_t &Gain )

Read Gains from gain table.

Parameters

Lin Input levels.

Gain Output gain.

The number of channels in Lin and Gain must match the number of bands times number of

channels in the gaintable.

5.86.3.5 unsigned int DynComp::gaintable_t::nbands ( ) const [inline]

Return number of frequency bands.

5.86.3.6 unsigned int DynComp::gaintable_t::nchannels ( ) const [inline]

Return number of audio channels.

5.86.3.7 std::vector<std::vector<mha_real_t > > gaintable_t::get_iofun ( ) const

Return current input-output function.

5.86.3.8 std::vector<mha_real_t>DynComp::gaintable_t::get_vL ( ) const [inline]

5.86.3.9 std::vector<mha_real_t>DynComp::gaintable_t::get_vF ( ) const [inline]

5.86.4 Member Data Documentation

5.86.4.1 unsigned int DynComp::gaintable_t::num_L [private]

5.86.4.2 unsigned int DynComp::gaintable_t::num_F [private]

5.86.4.3 unsigned int DynComp::gaintable_t::num_channels [private]

5.87 example1_t Class Reference 303

5.86.4.4 std::vector<mha_real_t>DynComp::gaintable_t::vL [private]

5.86.4.5 std::vector<mha_real_t>DynComp::gaintable_t::vF [private]

5.86.4.6 std::vector<mha_real_t>DynComp::gaintable_t::vFlog [private]

5.86.4.7 std::vector<std::vector<std::vector<mha_real_t> > > DynComp::gaintable_t::data

[private]

The documentation for this class was generated from the following ﬁles:

•gaintable.h

•gaintable.cpp

5.87 example1_t Class Reference

This C++ class implements the simplest example plugin for the step-by-step tutorial.

Inheritance diagram for example1_t:

example1_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

304 CONTENTS

Public Member Functions

•example1_t (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

Do-nothing constructor.

• void release (void)

Release may be empty.

• void prepare (mhaconﬁg_t &signal_info)

Plugin preparation.

•mha_wave_t ∗process (mha_wave_t ∗signal)

Signal processing performed by the plugin.

Additional Inherited Members

5.87.1 Detailed Description

This C++ class implements the simplest example plugin for the step-by-step tutorial.

It inherits from MHAPlugin::plugin_t (p. 728) for correct integration in the conﬁguration lan-

guage interface.

5.87.2 Constructor & Destructor Documentation

5.87.2.1 example1_t::example1_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )[inline]

Do-nothing constructor.

The constructor has to take these three arguments, but it does not have to use them. However,

the base class has to be initialized.

5.87.3 Member Function Documentation

5.87.3.1 void example1_t::release (

void ) [inline],[virtual]

Release may be empty.

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.87.3.2 void example1_t::prepare (

mhaconﬁg_t &signal_info )[inline],[virtual]

Plugin preparation.

This plugin checks that the input signal has the waveform domain and contains at least one

channel

5.88 example2_t Class Reference 305

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<int >(p. 730).

5.87.3.3 mha_wave_t∗example1_t::process (

mha_wave_t ∗signal )[inline]

Signal processing performed by the plugin.

This plugin multiplies the signal in the ﬁrst audio channel by a factor 0.1.

Parameters

signal Pointer to the input signal structure.

Returns

Returns a pointer to the input signal structure, with a the signal modiﬁed by this plugin.

(In-place processing)

The documentation for this class was generated from the following ﬁle:

•example1.cpp

5.88 example2_t Class Reference

This C++ class implements the second example plugin for the step-by-step tutorial.

306 CONTENTS

Inheritance diagram for example2_t:

example2_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•example2_t (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

• void prepare (mhaconﬁg_t &signal_info)

Plugin preparation.

• void release (void)

Undo restrictions posed in prepare.

•mha_wave_t ∗process (mha_wave_t ∗signal)

Signal processing performed by the plugin.

Private Attributes

•MHAParser::int_t scale_ch

Index of audio channel to scale.

•MHAParser::ﬂoat_t factor

The scaling factor applied to the selected channel.

5.88 example2_t Class Reference 307

Additional Inherited Members

5.88.1 Detailed Description

This C++ class implements the second example plugin for the step-by-step tutorial.

It extends the ﬁrst example by using conﬁguration language variables to inﬂuence the process-

ing.

5.88.2 Constructor & Destructor Documentation

5.88.2.1 example2_t::example2_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

5.88.3 Member Function Documentation

5.88.3.1 void example2_t::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

This plugin checks that the input signal has the waveform domain and contains enough chan-

nels.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<int >(p. 730).

5.88.3.2 void example2_t::release (

void ) [virtual]

Undo restrictions posed in prepare.

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

308 CONTENTS

5.88.3.3 mha_wave_t ∗example2_t::process (

mha_wave_t ∗signal )

Signal processing performed by the plugin.

This plugin multiplies the signal in the selected audio channel by the conﬁgured factor.

Parameters

signal Pointer to the input signal structure.

Returns

Returns a pointer to the input signal structure, with a the signal modiﬁed by this plugin.

(In-place processing)

5.88.4 Member Data Documentation

5.88.4.1 MHAParser::int_t example2_t::scale_ch [private]

Index of audio channel to scale.

5.88.4.2 MHAParser::ﬂoat_t example2_t::factor [private]

The scaling factor applied to the selected channel.

The documentation for this class was generated from the following ﬁle:

•example2.cpp

5.89 example3_t Class Reference

A Plugin class using the openMHA Event mechanism.

5.89 example3_t Class Reference 309

Inheritance diagram for example3_t:

example3_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•example3_t (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

• void prepare (mhaconﬁg_t &signal_info)

Plugin preparation.

• void release (void)

Bookkeeping only.

•mha_wave_t ∗process (mha_wave_t ∗signal)

Signal processing performed by the plugin.

Private Member Functions

• void on_scale_ch_writeaccess ()

• void on_scale_ch_valuechanged ()

• void on_scale_ch_readaccess ()

• void on_prereadaccess ()

310 CONTENTS

Private Attributes

•MHAParser::int_t scale_ch

Index of audio channel to scale.

•MHAParser::ﬂoat_t factor

The scaling factor applied to the selected channel.

•MHAParser::int_mon_t prepared

Keep Track of the prepare/release calls.

•MHAEvents::patchbay_t<example3_t >patchbay

The Event connector.

Additional Inherited Members

5.89.1 Detailed Description

A Plugin class using the openMHA Event mechanism.

This is the third example plugin for the step-by-step tutorial.

5.89.2 Constructor & Destructor Documentation

5.89.2.1 example3_t::example3_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

It connects the openMHA Events triggered by these conﬁguration variables to the respective

callbacks.

5.89.3 Member Function Documentation

5.89.3.1 void example3_t::on_scale_ch_writeaccess ( ) [private]

5.89.3.2 void example3_t::on_scale_ch_valuechanged ( ) [private]

5.89.3.3 void example3_t::on_scale_ch_readaccess ( ) [private]

5.89.3.4 void example3_t::on_prereadaccess ( ) [private]

5.89.3.5 void example3_t::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

This plugin checks that the input signal has the waveform domain and contains enough chan-

nels.

5.89 example3_t Class Reference 311

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<int >(p. 730).

5.89.3.6 void example3_t::release (

void ) [virtual]

Bookkeeping only.

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.89.3.7 mha_wave_t ∗example3_t::process (

mha_wave_t ∗signal )

Signal processing performed by the plugin.

This plugin multiplies the signal in the selected audio channel by the conﬁgured factor.

Parameters

signal Pointer to the input signal structure.

Returns

Returns a pointer to the input signal structure, with a the signal modiﬁed by this plugin.

(In-place processing)

5.89.4 Member Data Documentation

5.89.4.1 MHAParser::int_t example3_t::scale_ch [private]

Index of audio channel to scale.

5.89.4.2 MHAParser::ﬂoat_t example3_t::factor [private]

The scaling factor applied to the selected channel.

5.89.4.3 MHAParser::int_mon_t example3_t::prepared [private]

Keep Track of the prepare/release calls.

312 CONTENTS

5.89.4.4 MHAEvents::patchbay_t<example3_t>example3_t::patchbay [private]

The Event connector.

The documentation for this class was generated from the following ﬁle:

•example3.cpp

5.90 example4_t Class Reference

A Plugin class using the spectral signal.

Inheritance diagram for example4_t:

example4_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•example4_t (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

• void prepare (mhaconﬁg_t &signal_info)

Plugin preparation.

• void release (void)

Bookkeeping only.

•mha_spec_t ∗process (mha_spec_t ∗signal)

Signal processing performed by the plugin.

5.90 example4_t Class Reference 313

Private Member Functions

• void on_scale_ch_writeaccess ()

• void on_scale_ch_valuechanged ()

• void on_scale_ch_readaccess ()

• void on_prereadaccess ()

Private Attributes

•MHAParser::int_t scale_ch

Index of audio channel to scale.

•MHAParser::ﬂoat_t factor

The scaling factor applied to the selected channel.

•MHAParser::int_mon_t prepared

Keep Track of the prepare/release calls.

•MHAEvents::patchbay_t<example4_t >patchbay

The Event connector.

Additional Inherited Members

5.90.1 Detailed Description

A Plugin class using the spectral signal.

This is the fourth example plugin for the step-by-step tutorial.

5.90.2 Constructor & Destructor Documentation

5.90.2.1 example4_t::example4_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

This constructor initializes the conﬁguration language variables and inserts them into the

openMHA conﬁguration tree.

It connects the openMHA Events triggered by these conﬁguration variables to the respective

callbacks.

314 CONTENTS

5.90.3 Member Function Documentation

5.90.3.1 void example4_t::on_scale_ch_writeaccess ( ) [private]

5.90.3.2 void example4_t::on_scale_ch_valuechanged ( ) [private]

5.90.3.3 void example4_t::on_scale_ch_readaccess ( ) [private]

5.90.3.4 void example4_t::on_prereadaccess ( ) [private]

5.90.3.5 void example4_t::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

This plugin checks that the input signal has the waveform domain and contains enough chan-

nels.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<int >(p. 730).

5.90.3.6 void example4_t::release (

void ) [virtual]

Bookkeeping only.

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

5.90.3.7 mha_spec_t ∗example4_t::process (

mha_spec_t ∗signal )

Signal processing performed by the plugin.

This plugin multiplies the spectral signal in the selected audio channel by the conﬁgured factor.

Parameters

signal Pointer to the input signal structure.

5.91 example5_t Class Reference 315

Returns

Returns a pointer to the input signal structure, with a the signal modiﬁed by this plugin.

(In-place processing)

5.90.4 Member Data Documentation

5.90.4.1 MHAParser::int_t example4_t::scale_ch [private]

Index of audio channel to scale.

5.90.4.2 MHAParser::ﬂoat_t example4_t::factor [private]

The scaling factor applied to the selected channel.

5.90.4.3 MHAParser::int_mon_t example4_t::prepared [private]

Keep Track of the prepare/release calls.

5.90.4.4 MHAEvents::patchbay_t<example4_t>example4_t::patchbay [private]

The Event connector.

The documentation for this class was generated from the following ﬁle:

•example4.cpp

5.91 example5_t Class Reference

Public Member Functions

•example5_t (unsigned int, unsigned int, mha_real_t)

•mha_spec_t ∗process (mha_spec_t ∗)

Private Attributes

• unsigned int channel

•mha_real_t scale

316 CONTENTS

5.91.1 Constructor & Destructor Documentation

5.91.1.1 example5_t::example5_t (

unsigned int ichannel,

unsigned int numchannels,

mha_real_t iscale )

5.91.2 Member Function Documentation

5.91.2.1 mha_spec_t ∗example5_t::process (

mha_spec_t ∗spec )

5.91.3 Member Data Documentation

5.91.3.1 unsigned int example5_t::channel [private]

5.91.3.2 mha_real_t example5_t::scale [private]

The documentation for this class was generated from the following ﬁle:

•example5.cpp

5.92 example6_t Class Reference

Inheritance diagram for example6_t:

example6_t

MHAPlugin::plugin_t

< cfg_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< cfg_t >

5.92 example6_t Class Reference 317

Public Member Functions

•example6_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAParser::int_t channel_no

• ﬂoat rmsdb

•MHAEvents::patchbay_t<example6_t >patchbay

Additional Inherited Members

5.92.1 Constructor & Destructor Documentation

5.92.1.1 example6_t::example6_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.92.2 Member Function Documentation

5.92.2.1 mha_wave_t ∗example6_t::process (

mha_wave_t ∗wave )

5.92.2.2 void example6_t::prepare (

mhaconﬁg_t &tfcfg )[virtual]

Implements MHAPlugin::plugin_t<cfg_t >(p. 730).

5.92.2.3 void example6_t::update_cfg ( ) [private]

5.92.3 Member Data Documentation

5.92.3.1 MHAParser::int_t example6_t::channel_no [private]

5.92.3.2 ﬂoat example6_t::rmsdb [private]

5.92.3.3 MHAEvents::patchbay_t<example6_t>example6_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•example6.cpp

318 CONTENTS

5.93 expression_t Class Reference

Class for separating a string into a left hand value and a right hand value.

5.93.1 Detailed Description

Class for separating a string into a left hand value and a right hand value.

A list of valid operators can be provided. After construction, the class members lval, rval and

op contain the apropriate contents.

The documentation for this class was generated from the following ﬁle:

•mha_parser.cpp

5.94 fader_if_t Class Reference

Inheritance diagram for fader_if_t:

fader_if_t

MHAPlugin::plugin_t

< spec_fader_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< spec_fader_t >

Public Member Functions

•fader_if_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_spec_t ∗process (mha_spec_t ∗)

• void prepare (mhaconﬁg_t &)

5.94 fader_if_t Class Reference 319

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<fader_if_t >patchbay

•MHAParser::ﬂoat_t tau

•MHAParser::vﬂoat_t newgains

•mha_real_t ∗actgains

Additional Inherited Members

5.94.1 Constructor & Destructor Documentation

5.94.1.1 fader_if_t::fader_if_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.94.2 Member Function Documentation

5.94.2.1 mha_spec_t ∗fader_if_t::process (

mha_spec_t ∗s)

5.94.2.2 void fader_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<spec_fader_t >(p. 730).

5.94.2.3 void fader_if_t::update_cfg (

void ) [private]

5.94.3 Member Data Documentation

5.94.3.1 MHAEvents::patchbay_t<fader_if_t>fader_if_t::patchbay [private]

5.94.3.2 MHAParser::ﬂoat_t fader_if_t::tau [private]

5.94.3.3 MHAParser::vﬂoat_t fader_if_t::newgains [private]

5.94.3.4 mha_real_t∗fader_if_t::actgains [private]

The documentation for this class was generated from the following ﬁle:

•fader_spec.cpp

320 CONTENTS

5.95 fader_wave::fader_wave_if_t Class Reference

Inheritance diagram for fader_wave::fader_wave_if_t:

fader_wave::fader_wave_if_t

level_adaptor

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< runtime_cfg_t >

Public Member Functions

•fader_wave_if_t (algo_comm_t, const char ∗, const char ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void set_level ()

Private Attributes

•MHAParser::vﬂoat_t gain

•MHAParser::ﬂoat_t ramplen

•MHAEvents::patchbay_t<fader_wave_if_t >patchbay

• bool prepared

5.95 fader_wave::fader_wave_if_t Class Reference 321

Additional Inherited Members

5.95.1 Constructor & Destructor Documentation

5.95.1.1 fader_wave::fader_wave_if_t::fader_wave_if_t (

algo_comm_t iac,

const char ∗,

const char ∗)

5.95.2 Member Function Documentation

5.95.2.1 mha_wave_t ∗fader_wave::fader_wave_if_t::process (

mha_wave_t ∗s)

5.95.2.2 void fader_wave::fader_wave_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<runtime_cfg_t >(p. 730).

5.95.2.3 void fader_wave::fader_wave_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<runtime_cfg_t >(p. 731).

5.95.2.4 void fader_wave::fader_wave_if_t::set_level ( ) [private]

5.95.3 Member Data Documentation

5.95.3.1 MHAParser::vﬂoat_t fader_wave::fader_wave_if_t::gain [private]

5.95.3.2 MHAParser::ﬂoat_t fader_wave::fader_wave_if_t::ramplen [private]

5.95.3.3 MHAEvents::patchbay_t<fader_wave_if_t>fader_wave::fader_wave_if_t::patchbay

[private]

5.95.3.4 bool fader_wave::fader_wave_if_t::prepared [private]

The documentation for this class was generated from the following ﬁle:

•fader_wave.cpp

322 CONTENTS

5.96 fader_wave::level_adapt_t Class Reference

Inheritance diagram for fader_wave::level_adapt_t:

fader_wave::level_adapt_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•level_adapt_t (mhaconﬁg_t cf, mha_real_t adapt_len, std::vector<ﬂoat >l_new_←-

, std::vector<ﬂoat >l_old_)

• void update_frame ()

• std::vector<ﬂoat >get_level () const

• bool can_update () const

Private Attributes

• unsigned int ilen

• unsigned int pos

•MHAWindow::fun_t wnd

• std::vector<ﬂoat >l_new

• std::vector<ﬂoat >l_old

Additional Inherited Members

5.96.1 Constructor & Destructor Documentation

5.96.1.1 fader_wave::level_adapt_t::level_adapt_t (

mhaconﬁg_t cf,

mha_real_t adapt_len,

std::vector<ﬂoat >l_new_,

std::vector<ﬂoat >l_old_ )

5.97 fftfbpow::fftfbpow_interface_t Class Reference 323

5.96.2 Member Function Documentation

5.96.2.1 void fader_wave::level_adapt_t::update_frame ( )

5.96.2.2 std::vector<ﬂoat>fader_wave::level_adapt_t::get_level ( ) const [inline]

5.96.2.3 bool fader_wave::level_adapt_t::can_update ( ) const [inline]

5.96.3 Member Data Documentation

5.96.3.1 unsigned int fader_wave::level_adapt_t::ilen [private]

5.96.3.2 unsigned int fader_wave::level_adapt_t::pos [private]

5.96.3.3 MHAWindow::fun_t fader_wave::level_adapt_t::wnd [private]

5.96.3.4 std::vector<ﬂoat>fader_wave::level_adapt_t::l_new [private]

5.96.3.5 std::vector<ﬂoat>fader_wave::level_adapt_t::l_old [private]

The documentation for this class was generated from the following ﬁle:

•fader_wave.cpp

5.97 fftfbpow::fftfbpow_interface_t Class Reference

Interface class for fftfbpow plugin.

Inheritance diagram for fftfbpow::fftfbpow_interface_t:

fftfbpow::fftfbpow

_interface_t

MHAPlugin::plugin_t

< fftfbpow_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< fftfbpow_t >

MHAOvlFilter::fftfb

_vars_t

324 CONTENTS

Public Member Functions

•fftfbpow_interface_t (const algo_comm_t &ac, const std::string &, const std::string

&algo_name)

Constructor with standard MHA constructor parameters.

• void prepare (mhaconﬁg_t &tf)

Standard MHA plugin prepare function.

•mha_spec_t ∗process (mha_spec_t ∗s)

Standard MHA plugin process fct.

Private Member Functions

• void update_cfg ()

Constructs new runtime conﬁguration in thread-safe manner.

Private Attributes

• std::string name

Conﬁgured name of this plugin instance.

•MHAEvents::patchbay_t<fftfbpow_interface_t >patchbay

Patchbay to connect to MHA conﬁguration interface.

Additional Inherited Members

5.97.1 Detailed Description

Interface class for fftfbpow plugin.

5.97.2 Constructor & Destructor Documentation

5.97.2.1 fftfbpow::fftfbpow_interface_t::fftfbpow_interface_t (

const algo_comm_t &ac,

const std::string & ,

const std::string & algo_name )

Constructor with standard MHA constructor parameters.

Parameters

ac Handle to algorithm communication variable space

algo_name Conﬁgured name of this plugin instance

5.97 fftfbpow::fftfbpow_interface_t Class Reference 325

5.97.3 Member Function Documentation

5.97.3.1 void fftfbpow::fftfbpow_interface_t::prepare (

mhaconﬁg_t &tf )[virtual]

Standard MHA plugin prepare function.

Ensures that the input is in the frequency domain, calls update_cfg() (p. 325) and inserts fbpow

into the AC space.

Parameters

tf Incoming mha conﬁguration structure, contains information about input signal

Implements MHAPlugin::plugin_t<fftfbpow_t >(p. 730).

5.97.3.2 mha_spec_t ∗fftfbpow::fftfbpow_interface_t::process (

mha_spec_t ∗s)

Standard MHA plugin process fct.

Polls new conﬁg and calls process() (p. 325) of the runtime conﬁguration.

Parameters

sInput spectrum

Returns

Unchanged input spectrum

5.97.3.3 void fftfbpow::fftfbpow_interface_t::update_cfg (

void ) [private]

Constructs new runtime conﬁguration in thread-safe manner.

5.97.4 Member Data Documentation

5.97.4.1 std::string fftfbpow::fftfbpow_interface_t::name [private]

Conﬁgured name of this plugin instance.

326 CONTENTS

5.97.4.2 MHAEvents::patchbay_t<fftfbpow_interface_t>fftfbpow::fftfbpow_interface_t←-

::patchbay [private]

Patchbay to connect to MHA conﬁguration interface.

The documentation for this class was generated from the following ﬁle:

•fftfbpow.cpp

5.98 fftfbpow::fftfbpow_t Class Reference

Run time conﬁguration for the fftfbpow plugin.

Inheritance diagram for fftfbpow::fftfbpow_t:

fftfbpow::fftfbpow_t

MHAOvlFilter::fftfb_t

MHAOvlFilter::fspacing_t MHASignal::waveform_t

mha_wave_t

Public Member Functions

•fftfbpow_t (MHAOvlFilter::fftfb_vars_t &vars, unsigned int nch, unsigned int nfft, mha←-

_real_t fs, algo_comm_t ac, std::string name)

Constructor of the run time conﬁguration.

Public Attributes

•MHA_AC::waveform_t fbpow

AC variable containing the estimated power in each frequency band.

5.98 fftfbpow::fftfbpow_t Class Reference 327

Additional Inherited Members

5.98.1 Detailed Description

Run time conﬁguration for the fftfbpow plugin.

5.98.2 Constructor & Destructor Documentation

5.98.2.1 fftfbpow::fftfbpow_t::fftfbpow_t (

MHAOvlFilter::fftfb_vars_t &vars,

unsigned int nch,

unsigned int nfft,

mha_real_t fs,

algo_comm_t ac,

std::string name )

Constructor of the run time conﬁguration.

Parameters

vars Set of conﬁguration variables for FFT-based overlapping ﬁlters

nch Number of audio input channels

nfft Length of FFT

fs Sampling rate

ac AC space

name Conﬁgured name of plugin interface, used as preﬁx for AC variable names

5.98.3 Member Data Documentation

5.98.3.1 MHA_AC::waveform_t fftfbpow::fftfbpow_t::fbpow

AC variable containing the estimated power in each frequency band.

The documentation for this class was generated from the following ﬁle:

•fftfbpow.cpp

328 CONTENTS

5.99 fftﬁlterbank::fftfb_interface_t Class Reference

Inheritance diagram for fftﬁlterbank::fftfb_interface_t:

fftfilterbank::fftfb

_interface_t

MHAPlugin::plugin_t

< fftfb_plug_t >

MHAParser::parser_tMHAParser::base_t

MHAPlugin::config_t

< fftfb_plug_t >

MHAOvlFilter::overlap

_save_filterbank_t::vars_t

MHAOvlFilter::fftfb

_vars_t

Public Member Functions

•fftfb_interface_t (const algo_comm_t &ac, const std::string &th, const std::string &al)

Default values are set and MHA conﬁguration variables registered into the parser.

• void prepare (mhaconﬁg_t &)

Prepare all variables for processing.

• void release ()

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAParser::bool_t return_imag

•MHAEvents::patchbay_t<fftfb_interface_t >patchbay

•MHA_AC::int_t nchannels

• std::string algo

• bool prepared

• unsigned int nbands

Additional Inherited Members

5.99.1 Constructor & Destructor Documentation

5.99.1.1 fftﬁlterbank::fftfb_interface_t::fftfb_interface_t (

const algo_comm_t &ac,

const std::string & th,

const std::string & al )

Default values are set and MHA conﬁguration variables registered into the parser.

5.99 fftﬁlterbank::fftfb_interface_t Class Reference 329

Parameters

ac algorithm communication handle

th chain name

al algorithm name

5.99.2 Member Function Documentation

5.99.2.1 void fftﬁlterbank::fftfb_interface_t::prepare (

mhaconﬁg_t &tf )[virtual]

Prepare all variables for processing.

In this function, all variables are initialised and the ﬁlter shapes for each band are calculated.

The ﬁlter shapes W(f)are deﬁned as

W(f) = W(T(S(f))) = W(x), x =T(S(f)) = T(ˆ

f),

W(x)beeing a symmetric window function in the interval [−1,1] and S(f)the transformation

from the linear scale to the given frequency scale (see functions in FreqScaleFun). The function

T(ˆ

f)transforms the frequency range between the center frequencies [ˆ

fk−1,ˆ

fk]and [ˆ

fk,ˆ

fk+1]

into the interval [−1,0] and [0,1], respectively. This function is realised by the function linscale().

Parameters

tf Channel conﬁguration

Implements MHAPlugin::plugin_t<fftfb_plug_t >(p. 730).

5.99.2.2 void fftﬁlterbank::fftfb_interface_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<fftfb_plug_t >(p. 731).

5.99.2.3 mha_spec_t ∗fftﬁlterbank::fftfb_interface_t::process (

mha_spec_t ∗s)

5.99.2.4 mha_wave_t ∗fftﬁlterbank::fftfb_interface_t::process (

mha_wave_t ∗s)

5.99.2.5 void fftﬁlterbank::fftfb_interface_t::update_cfg (

void ) [private]

330 CONTENTS

5.99.3 Member Data Documentation

5.99.3.1 MHAParser::bool_t fftﬁlterbank::fftfb_interface_t::return_imag [private]

5.99.3.2 MHAEvents::patchbay_t<fftfb_interface_t>fftﬁlterbank::fftfb_interface_t::patchbay

[private]

5.99.3.3 MHA_AC::int_t fftﬁlterbank::fftfb_interface_t::nchannels [private]

5.99.3.4 std::string fftﬁlterbank::fftfb_interface_t::algo [private]

5.99.3.5 bool fftﬁlterbank::fftfb_interface_t::prepared [private]

5.99.3.6 unsigned int fftﬁlterbank::fftfb_interface_t::nbands [private]

The documentation for this class was generated from the following ﬁle:

•fftﬁlterbank.cpp

5.100 fftﬁlterbank::fftfb_plug_t Class Reference

Inheritance diagram for fftﬁlterbank::fftfb_plug_t:

fftfilterbank::fftfb

_plug_t

MHAOvlFilter::overlap

_save_filterbank_analytic_t

MHAOvlFilter::overlap

_save_filterbank_t

MHAOvlFilter::fftfb_t

MHAOvlFilter::fspacing_t MHASignal::waveform_t

mha_wave_t

MHAFilter::fftfilterbank_t

5.100 fftﬁlterbank::fftfb_plug_t Class Reference 331

Public Member Functions

•fftfb_plug_t (MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t &, mhaconﬁg_←-

tchcfg, algo_comm_t ac, std::string alg, bool return_imag)

•mha_spec_t ∗process (mha_spec_t ∗)

•mha_wave_t ∗process (mha_wave_t ∗)

• void insert ()

Private Attributes

•MHAOvlFilter::fftfb_ac_info_t fb_acinfo

•MHASignal::spectrum_t s_out

•MHA_AC::waveform_t imag

• bool return_imag_

Additional Inherited Members

5.100.1 Constructor & Destructor Documentation

5.100.1.1 fftﬁlterbank::fftfb_plug_t::fftfb_plug_t (

MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t &vars,

mhaconﬁg_t chcfg,

algo_comm_t ac,

std::string alg,

bool return_imag )

5.100.2 Member Function Documentation

5.100.2.1 mha_spec_t ∗fftﬁlterbank::fftfb_plug_t::process (

mha_spec_t ∗s)

5.100.2.2 mha_wave_t ∗fftﬁlterbank::fftfb_plug_t::process (

mha_wave_t ∗s)

5.100.2.3 void fftﬁlterbank::fftfb_plug_t::insert ( )

5.100.3 Member Data Documentation

5.100.3.1 MHAOvlFilter::fftfb_ac_info_t fftﬁlterbank::fftfb_plug_t::fb_acinfo [private]

5.100.3.2 MHASignal::spectrum_t fftﬁlterbank::fftfb_plug_t::s_out [private]

5.100.3.3 MHA_AC::waveform_t fftﬁlterbank::fftfb_plug_t::imag [private]

5.100.3.4 bool fftﬁlterbank::fftfb_plug_t::return_imag_ [private]

The documentation for this class was generated from the following ﬁle:

•fftﬁlterbank.cpp

332 CONTENTS

5.101 fshift::fshift_conﬁg_t Class Reference

fshift runtime conﬁg class

Public Member Functions

•fshift_conﬁg_t (fshift_t const ∗const plug)

C'tor of the fshift plugin runtime conﬁguration class.

•∼fshift_conﬁg_t ()=default

•mha_spec_t ∗process (mha_spec_t ∗)

Private Attributes

• const unsigned int kmin

FFT bin corresponding to fmin.

• const unsigned kmax

FFT bin corresponding to fmax.

• const int df

Frequency shift expressed in FFT bins.

• const mha_complex_t delta_phi

Phase advance per fft frame.

•mha_complex_t delta_phi_total

Sum of all phase advances.

5.101.1 Detailed Description

fshift runtime conﬁg class

5.101.2 Constructor & Destructor Documentation

5.101.2.1 fshift::fshift_conﬁg_t::fshift_conﬁg_t (

fshift_t const ∗const plug )[explicit]

C'tor of the fshift plugin runtime conﬁguration class.

Parameters

plug ptr to the plugin interface class. Conﬁguration information is given this way to keep the

argument list small.

5.101 fshift::fshift_conﬁg_t Class Reference 333

5.101.2.2 fshift::fshift_conﬁg_t::∼fshift_conﬁg_t ( ) [default]

5.101.3 Member Function Documentation

5.101.3.1 mha_spec_t ∗fshift::fshift_conﬁg_t::process (

mha_spec_t ∗in )

5.101.4 Member Data Documentation

5.101.4.1 const unsigned int fshift::fshift_conﬁg_t::kmin [private]

FFT bin corresponding to fmin.

5.101.4.2 const unsigned fshift::fshift_conﬁg_t::kmax [private]

FFT bin corresponding to fmax.

5.101.4.3 const int fshift::fshift_conﬁg_t::df [private]

Frequency shift expressed in FFT bins.

5.101.4.4 const mha_complex_t fshift::fshift_conﬁg_t::delta_phi [private]

Phase advance per fft frame.

5.101.4.5 mha_complex_t fshift::fshift_conﬁg_t::delta_phi_total [private]

Sum of all phase advances.

The documentation for this class was generated from the following ﬁles:

•fshift.hh

•fshift.cpp

334 CONTENTS

5.102 fshift::fshift_t Class Reference

fshift plugin interface class

Inheritance diagram for fshift::fshift_t:

fshift::fshift_t

MHAPlugin::plugin_t

< fshift_config_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< fshift_config_t >

Public Member Functions

•fshift_t (algo_comm_t &ac, const std::string &chain_name, const std::string &algo_←-

name)

Constructs our plugin.

•∼fshift_t ()

•mha_spec_t ∗process (mha_spec_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

•mha_real_t fmin () const

•mha_real_t fmax () const

•mha_real_t df () const

Private Member Functions

• void update_cfg ()

5.102 fshift::fshift_t Class Reference 335

Private Attributes

•MHAEvents::patchbay_t<fshift_t >patchbay

patch bay for connecting conﬁguration parser events with local member functions:

•MHAParser::ﬂoat_t m_fmin

•MHAParser::ﬂoat_t m_fmax

upper boundary for frequency shifter

•MHAParser::ﬂoat_t m_df

Shift frequency in Hz.

Additional Inherited Members

5.102.1 Detailed Description

fshift plugin interface class

5.102.2 Constructor & Destructor Documentation

5.102.2.1 fshift::fshift_t::fshift_t (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.102.2.2 fshift::fshift_t::∼fshift_t ( )

5.102.3 Member Function Documentation

5.102.3.1 mha_spec_t ∗fshift::fshift_t::process (

mha_spec_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.102.3.2 void fshift::fshift_t::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

336 CONTENTS

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<fshift_conﬁg_t >(p. 730).

5.102.3.3 void fshift::fshift_t::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<fshift_conﬁg_t >(p. 731).

5.102.3.4 mha_real_t fshift::fshift_t::fmin ( ) const [inline]

5.102.3.5 mha_real_t fshift::fshift_t::fmax ( ) const [inline]

5.102.3.6 mha_real_t fshift::fshift_t::df ( ) const [inline]

5.102.3.7 void fshift::fshift_t::update_cfg (

void ) [private]

5.102.4 Member Data Documentation

5.102.4.1 MHAEvents::patchbay_t<fshift_t>fshift::fshift_t::patchbay [private]

patch bay for connecting conﬁguration parser events with local member functions:

5.102.4.2 MHAParser::ﬂoat_t fshift::fshift_t::m_fmin [private]

5.102.4.3 MHAParser::ﬂoat_t fshift::fshift_t::m_fmax [private]

upper boundary for frequency shifter

5.102.4.4 MHAParser::ﬂoat_t fshift::fshift_t::m_df [private]

Shift frequency in Hz.

The documentation for this class was generated from the following ﬁles:

•fshift.hh

•fshift.cpp

5.103 fshift_hilbert::frequency_translator_t Class Reference 337

5.103 fshift_hilbert::frequency_translator_t Class Reference

Inheritance diagram for fshift_hilbert::frequency_translator_t:

fshift_hilbert::frequency

_translator_t

MHAPlugin::plugin_t

< hilbert_shifter_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< hilbert_shifter_t >

Public Member Functions

•frequency_translator_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_spec_t ∗process (mha_spec_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void update ()

Private Attributes

•MHAEvents::patchbay_t<frequency_translator_t >patchbay

•MHAParser::vﬂoat_t df

Vector containing the shift frequencies in Hz.

•MHAParser::ﬂoat_t fmin

Lower boundary for frequency shifter.

338 CONTENTS

•MHAParser::ﬂoat_t fmax

Upper boundary for frequency shifter.

•MHAParser::int_t irslen

Maximum length of cut off ﬁlter response.

•MHAParser::kw_t phasemode

Mode of gain smoothing.

Additional Inherited Members

5.103.1 Constructor & Destructor Documentation

5.103.1.1 fshift_hilbert::frequency_translator_t::frequency_translator_t (

const algo_comm_t &iac,

const std::string & ith,

const std::string & ial )

5.103.2 Member Function Documentation

5.103.2.1 mha_spec_t ∗fshift_hilbert::frequency_translator_t::process (

mha_spec_t ∗s)

5.103.2.2 void fshift_hilbert::frequency_translator_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<hilbert_shifter_t >(p. 730).

5.103.2.3 void fshift_hilbert::frequency_translator_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<hilbert_shifter_t >(p. 731).

5.103.2.4 void fshift_hilbert::frequency_translator_t::update ( ) [private]

5.103.3 Member Data Documentation

5.103.3.1 MHAEvents::patchbay_t<frequency_translator_t>fshift_hilbert::frequency_←-

translator_t::patchbay [private]

5.103.3.2 MHAParser::vﬂoat_t fshift_hilbert::frequency_translator_t::df [private]

Vector containing the shift frequencies in Hz.

5.104 fshift_hilbert::hilbert_shifter_t Class Reference 339

5.103.3.3 MHAParser::ﬂoat_t fshift_hilbert::frequency_translator_t::fmin [private]

Lower boundary for frequency shifter.

5.103.3.4 MHAParser::ﬂoat_t fshift_hilbert::frequency_translator_t::fmax [private]

Upper boundary for frequency shifter.

5.103.3.5 MHAParser::int_t fshift_hilbert::frequency_translator_t::irslen [private]

Maximum length of cut off ﬁlter response.

5.103.3.6 MHAParser::kw_t fshift_hilbert::frequency_translator_t::phasemode [private]

Mode of gain smoothing.

The documentation for this class was generated from the following ﬁle:

•fshift_hilbert.cpp

5.104 fshift_hilbert::hilbert_shifter_t Class Reference

Inheritance diagram for fshift_hilbert::hilbert_shifter_t:

fshift_hilbert::hilbert

_shifter_t

MHASignal::spectrum_t

mha_spec_t

340 CONTENTS

Public Member Functions

•hilbert_shifter_t (unsigned int fftlen, unsigned int channels,mha_real_t srate, unsigned

int kmin, unsigned int kmax, std::vector<mha_real_t >dphi, unsigned int frameshift,

unsigned int maxirslen, unsigned int phasemode)

•∼hilbert_shifter_t ()

• void process (mha_spec_t ∗)

Private Attributes

•MHASignal::spectrum_t fullspec

Part of the spectrum to be frequency shifted.

•MHASignal::spectrum_t analytic

Analytic signal, deﬁned as a(t)=x(t)+i∗H(x(t))

•MHASignal::waveform_t shifted

The frequency shifted signal in the time domain.

•MHASignal::spectrum_t mixw_shift

Helper varaible containing the coefﬁcients used to split the spectrum.

•MHASignal::spectrum_t mixw_ref

Helper varaible containing the coefﬁcients used to split the spectrum.

• fftw_plan plan_spec2analytic

FFT plan for the transformation of fullspec into the time domain.

•mha_fft_t mhafft

MHA wrapper object for fftw.

•MHASignal::waveform_t df

Vector holding one delta f value for every channel.

• unsigned int kmin

FFT frame that corresponds to f_min.

• unsigned int kmax

FFT frame that corresponds to f_max.

• unsigned int frameshift

Total phase advance within one fragment.

• std::vector<mha_complex_t >delta_phi

Phase advance per frame.

• std::vector<mha_complex_t >delta_phi_total

Sum of all phase advances.

Additional Inherited Members

5.104.1 Constructor & Destructor Documentation

5.104 fshift_hilbert::hilbert_shifter_t Class Reference 341

5.104.1.1 fshift_hilbert::hilbert_shifter_t::hilbert_shifter_t (

unsigned int fftlen,

unsigned int channels,

mha_real_t srate,

unsigned int kmin,

unsigned int kmax,

std::vector<mha_real_t >dphi,

unsigned int frameshift,

unsigned int maxirslen,

unsigned int phasemode )

5.104.1.2 fshift_hilbert::hilbert_shifter_t::∼hilbert_shifter_t ( )

5.104.2 Member Function Documentation

5.104.2.1 void fshift_hilbert::hilbert_shifter_t::process (

mha_spec_t ∗s)

5.104.3 Member Data Documentation

5.104.3.1 MHASignal::spectrum_t fshift_hilbert::hilbert_shifter_t::fullspec [private]

Part of the spectrum to be frequency shifted.

5.104.3.2 MHASignal::spectrum_t fshift_hilbert::hilbert_shifter_t::analytic [private]

Analytic signal, deﬁned as a(t)=x(t)+i∗H(x(t))

5.104.3.3 MHASignal::waveform_t fshift_hilbert::hilbert_shifter_t::shifted [private]

The frequency shifted signal in the time domain.

5.104.3.4 MHASignal::spectrum_t fshift_hilbert::hilbert_shifter_t::mixw_shift [private]

Helper varaible containing the coefﬁcients used to split the spectrum.

Contains 1 for every fft bin to be frequency shifted, 0 for all others

5.104.3.5 MHASignal::spectrum_t fshift_hilbert::hilbert_shifter_t::mixw_ref [private]

Helper varaible containing the coefﬁcients used to split the spectrum.

Contains 0 for every fft bin to be frequency shifted, 1 for all others

342 CONTENTS

5.104.3.6 fftw_plan fshift_hilbert::hilbert_shifter_t::plan_spec2analytic [private]

FFT plan for the transformation of fullspec into the time domain.

5.104.3.7 mha_fft_t fshift_hilbert::hilbert_shifter_t::mhafft [private]

MHA wrapper object for fftw.

5.104.3.8 MHASignal::waveform_t fshift_hilbert::hilbert_shifter_t::df [private]

Vector holding one delta f value for every channel.

5.104.3.9 unsigned int fshift_hilbert::hilbert_shifter_t::kmin [private]

FFT frame that corresponds to f_min.

5.104.3.10 unsigned int fshift_hilbert::hilbert_shifter_t::kmax [private]

FFT frame that corresponds to f_max.

5.104.3.11 unsigned int fshift_hilbert::hilbert_shifter_t::frameshift [private]

Total phase advance within one fragment.

5.104.3.12 std::vector<mha_complex_t>fshift_hilbert::hilbert_shifter_t::delta_phi [private]

Phase advance per frame.

5.104.3.13 std::vector<mha_complex_t>fshift_hilbert::hilbert_shifter_t::delta_phi_total

[private]

Sum of all phase advances.

The documentation for this class was generated from the following ﬁle:

•fshift_hilbert.cpp

5.105 fw_t Class Reference 343

5.105 fw_t Class Reference

Inheritance diagram for fw_t:

fw_t

mhaserver_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•fw_t ()

•∼fw_t ()

• bool exit_request () const

Protected Attributes

• int proc_error

• int io_error

Private Types

Private Member Functions

• void prepare ()

preparation for processing

• void start ()

start of processing

• void stop ()

344 CONTENTS

stop/pause of processing

• void release ()

release of IO device

• void quit ()

controlled quit

• void stopped (int, int)

• void started ()

• int process (mha_wave_t ∗,mha_wave_t ∗∗)

• void exec_fw_command ()

• void load_proc_lib ()

• void load_io_lib ()

• void fw_sleep_cmd ()

• void fw_until_cmd ()

• void get_input_signal_dimension ()

• void async_read ()

• void async_poll_msg ()

• void get_parserstate ()

Static Private Member Functions

• static void stopped (void ∗h, int proc_err, int io_err)

• static void started (void ∗h)

• static int process (void ∗h, mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

Private Attributes

•fw_vars_t prepare_vars

•MHAParser::int_mon_t nchannels_out

•MHAParser::string_t proc_name

•MHAParser::string_t io_name

•MHAParser::bool_t exit_on_stop

•MHAParser::int_t fw_sleep

•MHAParser::string_t fw_until

•MHAParser::kw_t fw_cmd

•MHAParser::string_mon_t parserstate

•MHAParser::string_t errorlog

•MHAParser::string_t fatallog

•MHAParser::vstring_t plugins

•MHAParser::vstring_t plugin_paths

•MHAParser::bool_t dump_mha

•MHAParser::string_t inst_name

A variable for naming MHA instances.

•MHAKernel::algo_comm_class_t ac

•PluginLoader::mhapluginloader_t ∗proc_lib

•io_lib_t ∗io_lib

5.105 fw_t Class Reference 345

•mhaconﬁg_t cﬁn

•mhaconﬁg_t cfout

•state_t state

• bool b_exit_request

•MHAParser::string_mon_t proc_error_string

•MHAEvents::patchbay_t<fw_t >patchbay

Additional Inherited Members

5.105.1 Member Enumeration Documentation

5.105.1.1 enum fw_t::state_t [private]

Enumerator

fw_unprepared

fw_stopped

fw_starting

fw_running

fw_stopping

fw_exiting

5.105.2 Constructor & Destructor Documentation

5.105.2.1 fw_t::fw_t ( )

5.105.2.2 fw_t::∼fw_t ( )

5.105.3 Member Function Documentation

5.105.3.1 bool fw_t::exit_request ( ) const [inline]

5.105.3.2 void fw_t::prepare (

void ) [private]

preparation for processing

5.105.3.3 void fw_t::start ( ) [private]

start of processing

346 CONTENTS

5.105.3.4 void fw_t::stop ( ) [private]

stop/pause of processing

5.105.3.5 void fw_t::release ( ) [private]

release of IO device

5.105.3.6 void fw_t::quit ( ) [private]

controlled quit

5.105.3.7 static void fw_t::stopped (

void ∗h,

int proc_err,

int io_err )[inline],[static],[private]

5.105.3.8 static void fw_t::started (

void ∗h)[inline],[static],[private]

5.105.3.9 static int fw_t::process (

void ∗h,

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[inline],[static],[private]

5.105.3.10 void fw_t::stopped (

int proc_err,

int io_err )[private]

5.105.3.11 void fw_t::started ( ) [private]

5.105.3.12 int fw_t::process (

mha_wave_t ∗s_in,

mha_wave_t ∗∗ s_out )[private]

5.105.3.13 void fw_t::exec_fw_command ( ) [private]

5.105.3.14 void fw_t::load_proc_lib ( ) [private]

5.105.3.15 void fw_t::load_io_lib ( ) [private]

5.105.3.16 void fw_t::fw_sleep_cmd ( ) [private]

5.105.3.17 void fw_t::fw_until_cmd ( ) [private]

5.105 fw_t Class Reference 347

5.105.3.18 void fw_t::get_input_signal_dimension ( ) [private]

5.105.3.19 void fw_t::async_read ( ) [inline],[private]

5.105.3.20 void fw_t::async_poll_msg ( ) [private]

5.105.3.21 void fw_t::get_parserstate ( ) [private]

5.105.4 Member Data Documentation

5.105.4.1 fw_vars_t fw_t::prepare_vars [private]

5.105.4.2 MHAParser::int_mon_t fw_t::nchannels_out [private]

5.105.4.3 MHAParser::string_t fw_t::proc_name [private]

5.105.4.4 MHAParser::string_t fw_t::io_name [private]

5.105.4.5 MHAParser::bool_t fw_t::exit_on_stop [private]

5.105.4.6 MHAParser::int_t fw_t::fw_sleep [private]

5.105.4.7 MHAParser::string_t fw_t::fw_until [private]

5.105.4.8 MHAParser::kw_t fw_t::fw_cmd [private]

5.105.4.9 MHAParser::string_mon_t fw_t::parserstate [private]

5.105.4.10 MHAParser::string_t fw_t::errorlog [private]

5.105.4.11 MHAParser::string_t fw_t::fatallog [private]

5.105.4.12 MHAParser::vstring_t fw_t::plugins [private]

5.105.4.13 MHAParser::vstring_t fw_t::plugin_paths [private]

5.105.4.14 MHAParser::bool_t fw_t::dump_mha [private]

5.105.4.15 MHAParser::string_t fw_t::inst_name [private]

A variable for naming MHA instances.

348 CONTENTS

5.105.4.16 MHAKernel::algo_comm_class_t fw_t::ac [private]

5.105.4.17 PluginLoader::mhapluginloader_t∗fw_t::proc_lib [private]

5.105.4.18 io_lib_t∗fw_t::io_lib [private]

5.105.4.19 mhaconﬁg_t fw_t::cﬁn [private]

5.105.4.20 mhaconﬁg_t fw_t::cfout [private]

5.105.4.21 state_t fw_t::state [private]

5.105.4.22 bool fw_t::b_exit_request [private]

5.105.4.23 int fw_t::proc_error [protected]

5.105.4.24 int fw_t::io_error [protected]

5.105.4.25 MHAParser::string_mon_t fw_t::proc_error_string [private]

5.105.4.26 MHAEvents::patchbay_t<fw_t>fw_t::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•mhafw_lib.h

•mhafw_lib.cpp

5.106 fw_vars_t Class Reference

Public Member Functions

•fw_vars_t (MHAParser::parser_t &)

• void lock_srate_fragsize ()

• void lock_channels ()

• void unlock_srate_fragsize ()

• void unlock_channels ()

Public Attributes

•MHAParser::int_t pinchannels

•MHAParser::int_t pfragmentsize

•MHAParser::ﬂoat_t psrate

5.106 fw_vars_t Class Reference 349

5.106.1 Constructor & Destructor Documentation

5.106.1.1 fw_vars_t::fw_vars_t (

MHAParser::parser_t &p)

5.106.2 Member Function Documentation

5.106.2.1 void fw_vars_t::lock_srate_fragsize ( )

5.106.2.2 void fw_vars_t::lock_channels ( )

5.106.2.3 void fw_vars_t::unlock_srate_fragsize ( )

5.106.2.4 void fw_vars_t::unlock_channels ( )

5.106.3 Member Data Documentation

5.106.3.1 MHAParser::int_t fw_vars_t::pinchannels

5.106.3.2 MHAParser::int_t fw_vars_t::pfragmentsize

5.106.3.3 MHAParser::ﬂoat_t fw_vars_t::psrate

The documentation for this class was generated from the following ﬁles:

•mhafw_lib.h

•mhafw_lib.cpp

350 CONTENTS

5.107 gain::gain_if_t Class Reference

Inheritance diagram for gain::gain_if_t:

gain::gain_if_t

MHAPlugin::plugin_t

< scaler_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< scaler_t >

Public Member Functions

•gain_if_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Private Member Functions

• void update_gain ()

• void update_bbgain ()

• void update_minmax ()

Private Attributes

•MHAEvents::patchbay_t<gain_if_t >patchbay

•MHAParser::vﬂoat_t gains

•MHAParser::ﬂoat_t bbgain

•MHAParser::ﬂoat_t vmin

•MHAParser::ﬂoat_t vmax

5.107 gain::gain_if_t Class Reference 351

Additional Inherited Members

5.107.1 Constructor & Destructor Documentation

5.107.1.1 gain::gain_if_t::gain_if_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.107.2 Member Function Documentation

5.107.2.1 mha_wave_t ∗gain::gain_if_t::process (

mha_wave_t ∗s)

5.107.2.2 mha_spec_t ∗gain::gain_if_t::process (

mha_spec_t ∗s)

5.107.2.3 void gain::gain_if_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<scaler_t >(p. 730).

5.107.2.4 void gain::gain_if_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<scaler_t >(p. 731).

5.107.2.5 void gain::gain_if_t::update_gain ( ) [private]

5.107.2.6 void gain::gain_if_t::update_bbgain ( ) [private]

5.107.2.7 void gain::gain_if_t::update_minmax ( ) [private]

5.107.3 Member Data Documentation

5.107.3.1 MHAEvents::patchbay_t<gain_if_t>gain::gain_if_t::patchbay [private]

5.107.3.2 MHAParser::vﬂoat_t gain::gain_if_t::gains [private]

5.107.3.3 MHAParser::ﬂoat_t gain::gain_if_t::bbgain [private]

5.107.3.4 MHAParser::ﬂoat_t gain::gain_if_t::vmin [private]

5.107.3.5 MHAParser::ﬂoat_t gain::gain_if_t::vmax [private]

The documentation for this class was generated from the following ﬁle:

•gain.cpp

352 CONTENTS

5.108 gain::scaler_t Class Reference

Inheritance diagram for gain::scaler_t:

gain::scaler_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•scaler_t (const unsigned int &channels, const MHAParser::vﬂoat_t &gains)

Additional Inherited Members

5.108.1 Constructor & Destructor Documentation

5.108.1.1 gain::scaler_t::scaler_t (

const unsigned int & channels,

const MHAParser::vﬂoat_t &gains )

The documentation for this class was generated from the following ﬁle:

•gain.cpp

5.109 gtfb_analyzer::gtfb_analyzer_cfg_t Struct Reference

Conﬁguration for Gammatone Filterbank Analyzer.

5.109 gtfb_analyzer::gtfb_analyzer_cfg_t Struct Reference 353

Public Member Functions

• unsigned bands () const

Each band is split into this number of bands.

• unsigned channels () const

The number of separate audio channels.

• unsigned frames () const

The number of frames in one chunk.

•mha_complex_t ∗states (unsigned channel, unsigned band)

Returns pointer to ﬁlter states for that band.

•gtfb_analyzer_cfg_t (unsigned ch, unsigned frames, unsigned ord, const std::vector<

mha_complex_t >&_coeff, const std::vector<mha_complex_t >&_norm_phase)

Create a conﬁguration for Gammatone Filterbank Analyzer.

•∼gtfb_analyzer_cfg_t ()

•mha_complex_t &cvalue (unsigned frame, unsigned channel, unsigned band)

Public Attributes

• unsigned order

The order of the gammatone ﬁlters.

• std::vector<mha_complex_t >coeff

The complex coefﬁcients of the gammatone ﬁlter bands.

• std::vector<mha_complex_t >norm_phase

Combination of normalization and phase correction factor.

•mha_wave_t s_out

Storage for the (complex) output signal.

• std::vector<mha_complex_t >state

Storage for Filter state.

5.109.1 Detailed Description

Conﬁguration for Gammatone Filterbank Analyzer.

5.109.2 Constructor & Destructor Documentation

5.109.2.1 gtfb_analyzer::gtfb_analyzer_cfg_t::gtfb_analyzer_cfg_t (

unsigned ch,

unsigned frames,

unsigned ord,

const std::vector<mha_complex_t >&_coeff,

const std::vector<mha_complex_t >&_norm_phase )[inline]

Create a conﬁguration for Gammatone Filterbank Analyzer.

354 CONTENTS

Parameters

ch Number of Audio channels.

frames Number of Audio frames per chunk.

ord The order of the gammatone ﬁlters.

_coeff Complex gammatone ﬁlter coefﬁcients.

_norm_phase Normalization and phase correction factors.

5.109.2.2 gtfb_analyzer::gtfb_analyzer_cfg_t::∼gtfb_analyzer_cfg_t ( ) [inline]

5.109.3 Member Function Documentation

5.109.3.1 unsigned gtfb_analyzer::gtfb_analyzer_cfg_t::bands ( ) const [inline]

Each band is split into this number of bands.

5.109.3.2 unsigned gtfb_analyzer::gtfb_analyzer_cfg_t::channels ( ) const [inline]

The number of separate audio channels.

5.109.3.3 unsigned gtfb_analyzer::gtfb_analyzer_cfg_t::frames ( ) const [inline]

The number of frames in one chunk.

5.109.3.4 mha_complex_t∗gtfb_analyzer::gtfb_analyzer_cfg_t::states (

unsigned channel,

unsigned band )[inline]

Returns pointer to ﬁlter states for that band.

5.109.3.5 mha_complex_t& gtfb_analyzer::gtfb_analyzer_cfg_t::cvalue (

unsigned frame,

unsigned channel,

unsigned band )[inline]

5.109.4 Member Data Documentation

5.109.4.1 unsigned gtfb_analyzer::gtfb_analyzer_cfg_t::order

The order of the gammatone ﬁlters.

5.109.4.2 std::vector<mha_complex_t>gtfb_analyzer::gtfb_analyzer_cfg_t::coeff

The complex coefﬁcients of the gammatone ﬁlter bands.

5.110 gtfb_analyzer::gtfb_analyzer_t Class Reference 355

5.109.4.3 std::vector<mha_complex_t>gtfb_analyzer::gtfb_analyzer_cfg_t::norm_phase

Combination of normalization and phase correction factor.

5.109.4.4 mha_wave_t gtfb_analyzer::gtfb_analyzer_cfg_t::s_out

Storage for the (complex) output signal.

Each of the real input audio channels is split into frequency bands with complex time signal

output. The split complex time signal is again stored in a mha_wave_t (p. 496) buffer. Each

complex time signal is stored as adjacent real and imaginary channels. Complex output from

one source channel is stored in adjacent complex output channels.

Example: If the input has 2 channels ch0 ch1, and gtfb_analyzer (p. 88) splits into 3 bands b0

b1 b2, then the order of output channels in s_out is: ch0_b0_real ch0_b0_imag ch0_b1_real

ch0_b1_imag ch0_b2_real ch0_b2_imag ch1_b0_real ch1_b1_imag ch1_b1_real ch1_b1_←-

imag ch1_b2_real ch1_b2_imag

5.109.4.5 std::vector<mha_complex_t>gtfb_analyzer::gtfb_analyzer_cfg_t::state

Storage for Filter state.

Holds channels() (p. 354)∗bands() (p. 354)∗order complex ﬁlter states. Layout←-

: state[(bands() (p. 354)∗channel+band)∗order+stage]

The documentation for this struct was generated from the following ﬁle:

•gtfb_analyzer.cpp

5.110 gtfb_analyzer::gtfb_analyzer_t Class Reference

Gammatone Filterbank Analyzer Plugin.

356 CONTENTS

Inheritance diagram for gtfb_analyzer::gtfb_analyzer_t:

gtfb_analyzer::gtfb

_analyzer_t

MHAPlugin::plugin_t

< gtfb_analyzer_cfg_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< gtfb_analyzer_cfg_t >

Public Member Functions

•gtfb_analyzer_t (const algo_comm_t &, const std::string &thread_name, const std←-

::string &algo_name)

•mha_wave_t ∗process (mha_wave_t ∗)

• void prepare (mhaconﬁg_t &)

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<gtfb_analyzer_t >patchbay

• bool prepared

•MHAParser::int_t order

•MHAParser::vcomplex_t coeff

•MHAParser::vcomplex_t norm_phase

5.110 gtfb_analyzer::gtfb_analyzer_t Class Reference 357

Additional Inherited Members

5.110.1 Detailed Description

Gammatone Filterbank Analyzer Plugin.

5.110.2 Constructor & Destructor Documentation

5.110.2.1 gtfb_analyzer::gtfb_analyzer_t::gtfb_analyzer_t (

const algo_comm_t &iac,

const std::string & thread_name,

const std::string & algo_name )

5.110.3 Member Function Documentation

5.110.3.1 mha_wave_t ∗gtfb_analyzer::gtfb_analyzer_t::process (

mha_wave_t ∗s)

5.110.3.2 void gtfb_analyzer::gtfb_analyzer_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<gtfb_analyzer_cfg_t >(p. 730).

5.110.3.3 void gtfb_analyzer::gtfb_analyzer_t::update_cfg (

void ) [private]

5.110.4 Member Data Documentation

5.110.4.1 MHAEvents::patchbay_t<gtfb_analyzer_t>gtfb_analyzer::gtfb_analyzer_t::patchbay

[private]

5.110.4.2 bool gtfb_analyzer::gtfb_analyzer_t::prepared [private]

5.110.4.3 MHAParser::int_t gtfb_analyzer::gtfb_analyzer_t::order [private]

5.110.4.4 MHAParser::vcomplex_t gtfb_analyzer::gtfb_analyzer_t::coeff [private]

5.110.4.5 MHAParser::vcomplex_t gtfb_analyzer::gtfb_analyzer_t::norm_phase [private]

The documentation for this class was generated from the following ﬁle:

•gtfb_analyzer.cpp

358 CONTENTS

5.111 hanning_ramps_t Class Reference

Public Member Functions

•hanning_ramps_t (unsigned int, unsigned int)

•∼hanning_ramps_t ()

• void operator() (MHASignal::waveform_t &)

Private Attributes

• unsigned int len_a

• unsigned int len_b

•mha_real_t ∗ramp_a

•mha_real_t ∗ramp_b

5.111.1 Constructor & Destructor Documentation

5.111.1.1 hanning_ramps_t::hanning_ramps_t (

unsigned int la,

unsigned int lb )

5.111.1.2 hanning_ramps_t::∼hanning_ramps_t (

void )

5.111.2 Member Function Documentation

5.111.2.1 void hanning_ramps_t::operator() (

MHASignal::waveform_t &b)

5.111.3 Member Data Documentation

5.111.3.1 unsigned int hanning_ramps_t::len_a [private]

5.111.3.2 unsigned int hanning_ramps_t::len_b [private]

5.111.3.3 mha_real_t∗hanning_ramps_t::ramp_a [private]

5.111.3.4 mha_real_t∗hanning_ramps_t::ramp_b [private]

The documentation for this class was generated from the following ﬁle:

•spec2wave.cpp

5.112 identity_t Class Reference 359

5.112 identity_t Class Reference

Inheritance diagram for identity_t:

identity_t

MHAPlugin::plugin_t

< int >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< int >

Public Member Functions

•identity_t (const algo_comm_t &, const std::string &, const std::string &)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Additional Inherited Members

5.112.1 Constructor & Destructor Documentation

5.112.1.1 identity_t::identity_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.112.2 Member Function Documentation

360 CONTENTS

5.112.2.1 mha_wave_t ∗identity_t::process (

mha_wave_t ∗s)

5.112.2.2 mha_spec_t ∗identity_t::process (

mha_spec_t ∗s)

5.112.2.3 void identity_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<int >(p. 730).

5.112.2.4 void identity_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<int >(p. 731).

The documentation for this class was generated from the following ﬁle:

•identity.cpp

5.113 iirﬁlter_t Class Reference

Inheritance diagram for iirﬁlter_t:

iirfilter_t

MHAFilter::iir_filter_t

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< iir_filter_state_t >

5.114 io_alsa_t Class Reference 361

Public Member Functions

•iirﬁlter_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare_ (mhaconﬁg_t &)

• void release_ ()

•mha_wave_t ∗process (mha_wave_t ∗)

Additional Inherited Members

5.113.1 Constructor & Destructor Documentation

5.113.1.1 iirﬁlter_t::iirﬁlter_t (

const algo_comm_t &ac,

const std::string & th,

const std::string & al )

5.113.2 Member Function Documentation

5.113.2.1 void iirﬁlter_t::prepare_ (

mhaconﬁg_t &tf )

5.113.2.2 void iirﬁlter_t::release_ ( ) [inline]

5.113.2.3 mha_wave_t ∗iirﬁlter_t::process (

mha_wave_t ∗s)

The documentation for this class was generated from the following ﬁle:

•iirﬁlter.cpp

5.114 io_alsa_t Class Reference

MHA IO interface class for ALSA IO.

Inheritance diagram for io_alsa_t:

io_alsa_t

MHAParser::parser_t

MHAParser::base_t

362 CONTENTS

Public Member Functions

•io_alsa_t (unsigned int fragsize, ﬂoat samplerate, IOProcessEvent_t proc_event, void

∗proc_handle,IOStartedEvent_t start_event, void ∗start_handle,IOStoppedEvent←-

_t stop_event, void ∗stop_handle)

Constructor, receives the callback handles to interact with the MHA framework.

•template<typename T = void>

void prepare (int, int)

Called after the framework has perpared the processing plugins and the number of input and

output channels are ﬁxed.

• void release ()

MHA framework leaves prepared state.

• void start ()

MHA framework calls this function when signal processing should start.

• void stop ()

MHA framework calls this function when signal processing should stop.

•template<>

void prepare (int nch_in, int nch_out)

Static Public Member Functions

• static void ∗thread_start (void ∗h)

MHAIOAlsa uses a separate thread that calls the alsa read and write functions to read and

write audio samples, these functions are blocking until samples can be read or written.

Private Member Functions

• void process ()

Private Attributes

• bool b_process

• unsigned int fw_fragsize

• unsigned int fw_samplerate

•IOProcessEvent_t proc_event

• void ∗proc_handle

•IOStartedEvent_t start_event

• void ∗start_handle

•IOStoppedEvent_t stop_event

• void ∗stop_handle

•alsa_base_t ∗dev_in

•alsa_base_t ∗dev_out

• pthread_t proc_thread

•alsa_dev_par_parser_t p_in

•alsa_dev_par_parser_t p_out

•MHAParser::bool_t pcmlink

•MHAParser::int_t priority

•MHAParser::kw_t format

•MHAParser::int_mon_t alsa_start_counter

•MHAEvents::patchbay_t<io_alsa_t >patchbay

5.114 io_alsa_t Class Reference 363

Additional Inherited Members

5.114.1 Detailed Description

MHA IO interface class for ALSA IO.

5.114.2 Constructor & Destructor Documentation

5.114.2.1 io_alsa_t::io_alsa_t (

unsigned int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

Constructor, receives the callback handles to interact with the MHA framework.

5.114.3 Member Function Documentation

5.114.3.1 template<typename T >void io_alsa_t::prepare (

int nch_in,

int nch_out )

Called after the framework has perpared the processing plugins and the number of input and

output channels are ﬁxed.

open pcm streams

5.114.3.2 void io_alsa_t::release (

void )

MHA framework leaves prepared state.

5.114.3.3 void io_alsa_t::start ( )

MHA framework calls this function when signal processing should start.

5.114.3.4 void io_alsa_t::stop ( )

MHA framework calls this function when signal processing should stop.

364 CONTENTS

5.114.3.5 void ∗io_alsa_t::thread_start (

void ∗h)[static]

MHAIOAlsa uses a separate thread that calls the alsa read and write functions to read and

write audio samples, these functions are blocking until samples can be read or written.

This is the start function of that thread.

5.114.3.6 void io_alsa_t::process (

void ) [private]

5.114.3.7 template<> void io_alsa_t::prepare (

int nch_in,

int nch_out )

5.114.4 Member Data Documentation

5.114.4.1 bool io_alsa_t::b_process [private]

5.114.4.2 unsigned int io_alsa_t::fw_fragsize [private]

5.114.4.3 unsigned int io_alsa_t::fw_samplerate [private]

5.114.4.4 IOProcessEvent_t io_alsa_t::proc_event [private]

5.114.4.5 void∗io_alsa_t::proc_handle [private]

5.114.4.6 IOStartedEvent_t io_alsa_t::start_event [private]

5.114.4.7 void∗io_alsa_t::start_handle [private]

5.114.4.8 IOStoppedEvent_t io_alsa_t::stop_event [private]

5.114.4.9 void∗io_alsa_t::stop_handle [private]

5.114.4.10 alsa_base_t∗io_alsa_t::dev_in [private]

5.114.4.11 alsa_base_t∗io_alsa_t::dev_out [private]

5.114.4.12 pthread_t io_alsa_t::proc_thread [private]

5.114.4.13 alsa_dev_par_parser_t io_alsa_t::p_in [private]

5.114.4.14 alsa_dev_par_parser_t io_alsa_t::p_out [private]

5.114.4.15 MHAParser::bool_t io_alsa_t::pcmlink [private]

5.114.4.16 MHAParser::int_t io_alsa_t::priority [private]

5.114.4.17 MHAParser::kw_t io_alsa_t::format [private]

5.114.4.18 MHAParser::int_mon_t io_alsa_t::alsa_start_counter [private]

5.114.4.19 MHAEvents::patchbay_t<io_alsa_t>io_alsa_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOalsa.cpp

5.115 io_ﬁle_t Class Reference 365

5.115 io_ﬁle_t Class Reference

File IO.

Inheritance diagram for io_ﬁle_t:

io_file_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•io_ﬁle_t (int fragsize, ﬂoat samplerate,IOProcessEvent_t proc_event, void ∗proc_←-

handle,IOStartedEvent_t start_event, void ∗start_handle,IOStoppedEvent_t stop←-

_event, void ∗stop_handle)

•∼io_ﬁle_t ()

• void prepare (int, int)

Allocate buffers, activate FILE client and install internal ports.

• void start ()

• void stop ()

• void release ()

Remove FILE client and deallocate internal ports and buffers.

Private Member Functions

• void stopped (int, int)

• void setlock (bool locked)

lock or unlock all parser variables.

366 CONTENTS

Private Attributes

• int fragsize

• ﬂoat samplerate

• int nchannels_in

• int nchannels_ﬁle_in

• int nchannels_out

•IOProcessEvent_t proc_event

• void ∗proc_handle

•IOStartedEvent_t start_event

• void ∗start_handle

•IOStoppedEvent_t stop_event

• void ∗stop_handle

•MHAParser::string_t ﬁlename_input

•MHAParser::string_t ﬁlename_output

•MHAParser::kw_t output_sample_format

•MHAParser::int_t startsample

•MHAParser::int_t length

•MHAParser::bool_t strict_channel_match

•MHAParser::bool_t strict_srate_match

•MHASignal::waveform_t ∗s_in

•MHASignal::waveform_t ∗s_ﬁle_in

•mha_wave_t ∗s_out

• bool b_prepared

• SNDFILE ∗sf_in

• SNDFILE ∗sf_out

• SF_INFO sﬁnf_in

• SF_INFO sﬁnf_out

• sf_count_t total_read

Additional Inherited Members

5.115.1 Detailed Description

File IO.

5.115.2 Constructor & Destructor Documentation

5.115.2.1 io_ﬁle_t::io_ﬁle_t (

int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

5.115 io_ﬁle_t Class Reference 367

5.115.2.2 io_ﬁle_t::∼io_ﬁle_t ( )

5.115.3 Member Function Documentation

5.115.3.1 void io_ﬁle_t::prepare (

int nch_in,

int nch_out )

Allocate buffers, activate FILE client and install internal ports.

5.115.3.2 void io_ﬁle_t::start ( )

5.115.3.3 void io_ﬁle_t::stop ( )

5.115.3.4 void io_ﬁle_t::release (

void )

Remove FILE client and deallocate internal ports and buffers.

5.115.3.5 void io_ﬁle_t::stopped (

int proc_err,

int io_err )[private]

5.115.3.6 void io_ﬁle_t::setlock (

bool locked )[private]

lock or unlock all parser variables.

Used in prepare/release.

Parameters

locked When true, locks. When false, unlocks.

5.115.4 Member Data Documentation

5.115.4.1 int io_ﬁle_t::fragsize [private]

5.115.4.2 ﬂoat io_ﬁle_t::samplerate [private]

5.115.4.3 int io_ﬁle_t::nchannels_in [private]

5.115.4.4 int io_ﬁle_t::nchannels_ﬁle_in [private]

368 CONTENTS

5.115.4.5 int io_ﬁle_t::nchannels_out [private]

5.115.4.6 IOProcessEvent_t io_ﬁle_t::proc_event [private]

5.115.4.7 void∗io_ﬁle_t::proc_handle [private]

5.115.4.8 IOStartedEvent_t io_ﬁle_t::start_event [private]

5.115.4.9 void∗io_ﬁle_t::start_handle [private]

5.115.4.10 IOStoppedEvent_t io_ﬁle_t::stop_event [private]

5.115.4.11 void∗io_ﬁle_t::stop_handle [private]

5.115.4.12 MHAParser::string_t io_ﬁle_t::ﬁlename_input [private]

5.115.4.13 MHAParser::string_t io_ﬁle_t::ﬁlename_output [private]

5.115.4.14 MHAParser::kw_t io_ﬁle_t::output_sample_format [private]

5.115.4.15 MHAParser::int_t io_ﬁle_t::startsample [private]

5.115.4.16 MHAParser::int_t io_ﬁle_t::length [private]

5.115.4.17 MHAParser::bool_t io_ﬁle_t::strict_channel_match [private]

5.115.4.18 MHAParser::bool_t io_ﬁle_t::strict_srate_match [private]

5.115.4.19 MHASignal::waveform_t∗io_ﬁle_t::s_in [private]

5.115.4.20 MHASignal::waveform_t∗io_ﬁle_t::s_ﬁle_in [private]

5.115.4.21 mha_wave_t∗io_ﬁle_t::s_out [private]

5.115.4.22 bool io_ﬁle_t::b_prepared [private]

5.115.4.23 SNDFILE∗io_ﬁle_t::sf_in [private]

5.115.4.24 SNDFILE∗io_ﬁle_t::sf_out [private]

5.115.4.25 SF_INFO io_ﬁle_t::sﬁnf_in [private]

5.115.4.26 SF_INFO io_ﬁle_t::sﬁnf_out [private]

5.115.4.27 sf_count_t io_ﬁle_t::total_read [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOFile.cpp

5.116 io_lib_t Class Reference 369

5.116 io_lib_t Class Reference

Class for loading MHA sound IO module.

Inheritance diagram for io_lib_t:

io_lib_t

MHAParser::c_ifc_parser_t

MHAParser::base_t

Public Member Functions

•io_lib_t (int fragsize, ﬂoat samplerate, IOProcessEvent_t proc_event, void ∗proc_←-

handle, IOStartedEvent_t start_event, void ∗start_handle, IOStoppedEvent_t stop_←-

event, void ∗stop_handle, std::string libname)

load and initialize MHA sound io module.

•∼io_lib_t ()

Deinitialize and unload this MHA sound io module.

• void prepare (unsigned int inch, unsigned int outch)

Prepare the sound io module.

• void start ()

Tell the sound io module to start sound processing.

• void stop ()

• void release ()

• std::string lib_str_error (int err)

Protected Member Functions

• void test_error ()

370 CONTENTS

Protected Attributes

• int lib_err

•dynamiclib_t lib_handle

• void ∗lib_data

•IOInit_t IOInit_cb

•IOPrepare_t IOPrepare_cb

•IOStart_t IOStart_cb

•IOStop_t IOStop_cb

•IORelease_t IORelease_cb

•IOSetVar_t IOSetVar_cb

•IOStrError_t IOStrError_cb

•IODestroy_t IODestroy_cb

Additional Inherited Members

5.116.1 Detailed Description

Class for loading MHA sound IO module.

5.116.2 Constructor & Destructor Documentation

5.116.2.1 io_lib_t::io_lib_t (

int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle,

std::string libname )

load and initialize MHA sound io module.

5.116.2.2 io_lib_t::∼io_lib_t ( )

Deinitialize and unload this MHA sound io module.

5.116.3 Member Function Documentation

5.116.3.1 void io_lib_t::prepare (

unsigned int inch,

unsigned int outch )

Prepare the sound io module.

After preparation, the sound io module may start the sound processing at any time (external

trigger). When the sound processing is started, the sound io module will call the start_event

callback.

5.116 io_lib_t Class Reference 371

Parameters

inch number of input channels

outch number of output channels

5.116.3.2 void io_lib_t::start ( )

Tell the sound io module to start sound processing.

Some io modules need this, for others that wait for external events this method might do nothing.

5.116.3.3 void io_lib_t::stop ( )

5.116.3.4 void io_lib_t::release ( )

5.116.3.5 std::string io_lib_t::lib_str_error (

int err )

5.116.3.6 void io_lib_t::test_error ( ) [protected]

5.116.4 Member Data Documentation

5.116.4.1 int io_lib_t::lib_err [protected]

5.116.4.2 dynamiclib_t io_lib_t::lib_handle [protected]

5.116.4.3 void∗io_lib_t::lib_data [protected]

5.116.4.4 IOInit_t io_lib_t::IOInit_cb [protected]

5.116.4.5 IOPrepare_t io_lib_t::IOPrepare_cb [protected]

5.116.4.6 IOStart_t io_lib_t::IOStart_cb [protected]

5.116.4.7 IOStop_t io_lib_t::IOStop_cb [protected]

5.116.4.8 IORelease_t io_lib_t::IORelease_cb [protected]

5.116.4.9 IOSetVar_t io_lib_t::IOSetVar_cb [protected]

5.116.4.10 IOStrError_t io_lib_t::IOStrError_cb [protected]

5.116.4.11 IODestroy_t io_lib_t::IODestroy_cb [protected]

The documentation for this class was generated from the following ﬁles:

•mhafw_lib.h

•mhafw_lib.cpp

372 CONTENTS

5.117 io_parser_t Class Reference

Main class for Parser IO.

Inheritance diagram for io_parser_t:

io_parser_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•io_parser_t (unsigned int fragsize,IOProcessEvent_t proc_event, void ∗proc_handle,

IOStartedEvent_t start_event, void ∗start_handle,IOStoppedEvent_t stop_event,

void ∗stop_handle)

•∼io_parser_t ()

• void prepare (int, int)

Allocate buffers, activate JACK client and install internal ports.

• void start ()

• void stop ()

• void release ()

Remove JACK client and deallocate internal ports and buffers.

Private Member Functions

• void stopped (int, int)

• void started ()

• void process_frame ()

5.117 io_parser_t Class Reference 373

Private Attributes

• unsigned int fragsize

• unsigned int nchannels_in

• unsigned int nchannels_out

•IOProcessEvent_t proc_event

• void ∗proc_handle

•IOStartedEvent_t start_event

• void ∗start_handle

•IOStoppedEvent_t stop_event

• void ∗stop_handle

•MHAParser::mﬂoat_t input

•MHAParser::mﬂoat_mon_t output

•MHASignal::waveform_t ∗s_in

•mha_wave_t ∗s_out

• bool b_fw_started

• bool b_stopped

• bool b_prepared

• bool b_starting

•MHAEvents::patchbay_t<io_parser_t >patchbay

Additional Inherited Members

5.117.1 Detailed Description

Main class for Parser IO.

5.117.2 Constructor & Destructor Documentation

5.117.2.1 io_parser_t::io_parser_t (

unsigned int fragsize,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

5.117.2.2 io_parser_t::∼io_parser_t ( )

5.117.3 Member Function Documentation

5.117.3.1 void io_parser_t::prepare (

int nch_in,

int nch_out )

Allocate buffers, activate JACK client and install internal ports.

374 CONTENTS

5.117.3.2 void io_parser_t::start ( )

5.117.3.3 void io_parser_t::stop ( )

5.117.3.4 void io_parser_t::release (

void )

Remove JACK client and deallocate internal ports and buffers.

5.117.3.5 void io_parser_t::stopped (

int proc_err,

int io_err )[private]

5.117.3.6 void io_parser_t::started ( ) [private]

5.117.3.7 void io_parser_t::process_frame ( ) [private]

5.117.4 Member Data Documentation

5.117.4.1 unsigned int io_parser_t::fragsize [private]

5.117.4.2 unsigned int io_parser_t::nchannels_in [private]

5.117.4.3 unsigned int io_parser_t::nchannels_out [private]

5.117.4.4 IOProcessEvent_t io_parser_t::proc_event [private]

5.117.4.5 void∗io_parser_t::proc_handle [private]

5.117.4.6 IOStartedEvent_t io_parser_t::start_event [private]

5.117.4.7 void∗io_parser_t::start_handle [private]

5.117.4.8 IOStoppedEvent_t io_parser_t::stop_event [private]

5.117.4.9 void∗io_parser_t::stop_handle [private]

5.117.4.10 MHAParser::mﬂoat_t io_parser_t::input [private]

5.117.4.11 MHAParser::mﬂoat_mon_t io_parser_t::output [private]

5.117.4.12 MHASignal::waveform_t∗io_parser_t::s_in [private]

5.117.4.13 mha_wave_t∗io_parser_t::s_out [private]

5.117.4.14 bool io_parser_t::b_fw_started [private]

5.117.4.15 bool io_parser_t::b_stopped [private]

5.117.4.16 bool io_parser_t::b_prepared [private]

5.117.4.17 bool io_parser_t::b_starting [private]

5.117.4.18 MHAEvents::patchbay_t<io_parser_t>io_parser_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOParser.cpp

5.118 io_tcp_fwcb_t Class Reference 375

5.118 io_tcp_fwcb_t Class Reference

TCP sound-io library's interface to the framework callbacks.

Public Member Functions

•io_tcp_fwcb_t (IOProcessEvent_t proc_event, void ∗proc_handle,IOStartedEvent←-

_t start_event, void ∗start_handle,IOStoppedEvent_t stop_event, void ∗stop_←-

handle)

Constructor stores framework handles and initializes error numbers to 0.

• virtual ∼io_tcp_fwcb_t ()

Do-nothing destructor.

• virtual void start ()

Call the framework's start callback.

• virtual int process (mha_wave_t ∗sIn, mha_wave_t ∗&sOut)

Call the frameworks processing callback.

• virtual void set_errnos (int proc_err, int io_err)

Save error numbers to use during.

• virtual void stop ()

Call the frameworks stop callback.

Private Attributes

•IOProcessEvent_t proc_event

Pointer to signal processing callback function.

•IOStartedEvent_t start_event

Pointer to start notiﬁcation callback function.

•IOStoppedEvent_t stop_event

Pointer to stop notiﬁcation callback function.

• void ∗proc_handle

Handles belonging to framework.

• void ∗start_handle

• void ∗stop_handle

• int proc_err

Errors from the processing callback and from the TCP IO itself are stored here before closing

Network handles.

• int io_err

5.118.1 Detailed Description

TCP sound-io library's interface to the framework callbacks.

376 CONTENTS

5.118.2 Constructor & Destructor Documentation

5.118.2.1 io_tcp_fwcb_t::io_tcp_fwcb_t (

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

Constructor stores framework handles and initializes error numbers to 0.

5.118.2.2 virtual io_tcp_fwcb_t::∼io_tcp_fwcb_t ( ) [inline],[virtual]

Do-nothing destructor.

5.118.3 Member Function Documentation

5.118.3.1 void io_tcp_fwcb_t::start ( ) [virtual]

Call the framework's start callback.

5.118.3.2 int io_tcp_fwcb_t::process (

mha_wave_t ∗sIn,

mha_wave_t ∗&sOut )[virtual]

Call the frameworks processing callback.

Parameters

sIn The input sound data just received from TCP.

sOut A pointer to output sound data. Will point to the output sound data storage when the

callback ﬁnishes.

Returns

Status, an error number from the signal processing callback. If this is != 0, then the

connection should be closed.

5.118.3.3 void io_tcp_fwcb_t::set_errnos (

int proc_err,

int io_err )[virtual]

Save error numbers to use during.

5.118 io_tcp_fwcb_t Class Reference 377

See also

stop (p. 377)

Parameters

proc_err The error number from the

See also

process (p. 376) callback.

Parameters

io_err The error number from the io library itself.

5.118.3.4 void io_tcp_fwcb_t::stop ( ) [virtual]

Call the frameworks stop callback.

Uses the error numbers set previously with

See also

set_errnos (p. 376).

5.118.4 Member Data Documentation

5.118.4.1 IOProcessEvent_t io_tcp_fwcb_t::proc_event [private]

Pointer to signal processing callback function.

5.118.4.2 IOStartedEvent_t io_tcp_fwcb_t::start_event [private]

Pointer to start notiﬁcation callback function.

Called when a new TCP connection is established or the user issues the start command while

there is a connection.

5.118.4.3 IOStoppedEvent_t io_tcp_fwcb_t::stop_event [private]

Pointer to stop notiﬁcation callback function.

Called when the connection is closed.

378 CONTENTS

5.118.4.4 void∗io_tcp_fwcb_t::proc_handle [private]

Handles belonging to framework.

5.118.4.5 void ∗io_tcp_fwcb_t::start_handle [private]

5.118.4.6 void ∗io_tcp_fwcb_t::stop_handle [private]

5.118.4.7 int io_tcp_fwcb_t::proc_err [private]

Errors from the processing callback and from the TCP IO itself are stored here before closing

Network handles.

MHAIOTCP is notiﬁed by the server when the connection has been taken down, and calls

See also

stop (p. 377) from that callback. Within stop (p. 377), these error numbers are read again

and transmitted to the framework.

5.118.4.8 int io_tcp_fwcb_t::io_err [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOTCP.cpp

5.119 io_tcp_parser_t Class Reference

The parser interface of the IOTCP library.

Inheritance diagram for io_tcp_parser_t:

io_tcp_parser_t

MHAParser::parser_t

MHAParser::base_t

5.119 io_tcp_parser_t Class Reference 379

Public Member Functions

• virtual const std::string & get_local_address () const

Read parser variable local_address, this is the address of the network interface that should

listen for incoming connections.

• virtual unsigned short get_local_port () const

Read parser variable local_port, this is the TCP port that should be used for incoming connec-

tions.

• virtual void set_local_port (unsigned short port)

Set parser variable local_port.

• virtual bool get_server_port_open () const

Return the status of the server port as it is known to the parser.

• virtual void set_server_port_open (bool open)

Inform the parser of the current status of the server socket.

• virtual bool get_connected () const

Return the parser's knowledge concerning wether there currently exists an established sound

data TCP connection or not.

• virtual void set_connected (bool connected)

Inform the parser about the existance of a sound data connection.

• virtual void set_new_peer (unsigned short port, const std::string &host)

Set parser monitor variables peer_port and peer_address, and calls set_connected(true).

•io_tcp_parser_t ()

Constructor initializes parser variables.

• virtual ∼io_tcp_parser_t ()

Do-nothing destructor.

• virtual void debug (const std::string &message)

Private Attributes

•MHAParser::string_t local_address

Lets the user set the local network interface to listen on.

•MHAParser::int_t local_port

Lets the user choose the local tcp port to listen on.

•MHAParser::int_mon_t server_port_open

Indicates wether the TCP server socket is currently open.

•MHAParser::int_mon_t connected

Indicator if there currently is a sound data connection over TCP.

•MHAParser::string_mon_t peer_address

Display the ip address of the currently connected sound data client.

•MHAParser::int_mon_t peer_port

Display the tcp port used by the current sound data client.

•MHAParser::string_t debug_ﬁlename

ﬁlename to write debugging info to (if non-empty)

• FILE ∗debug_ﬁle

ﬁle handle to write debugging info to

380 CONTENTS

Additional Inherited Members

5.119.1 Detailed Description

The parser interface of the IOTCP library.

5.119.2 Constructor & Destructor Documentation

5.119.2.1 io_tcp_parser_t::io_tcp_parser_t ( )

Constructor initializes parser variables.

5.119.2.2 virtual io_tcp_parser_t::∼io_tcp_parser_t ( ) [inline],[virtual]

Do-nothing destructor.

5.119.3 Member Function Documentation

5.119.3.1 virtual const std::string& io_tcp_parser_t::get_local_address ( ) const [inline],

[virtual]

Read parser variable local_address, this is the address of the network interface that should

listen for incoming connections.

Returns

A string containing the address of the local interface as it was set by the user.

5.119.3.2 unsigned short io_tcp_parser_t::get_local_port ( ) const [virtual]

Read parser variable local_port, this is the TCP port that should be used for incoming connec-

tions.

Returns

The local tcp port to listen on as it was chosen by the user. The port number is between

MIN_TCP_PORT and MAX_TCP_PORT.

5.119.3.3 void io_tcp_parser_t::set_local_port (

unsigned short port )[virtual]

Set parser variable local_port.

This is needed when it was set to 0 before: In this case, the OS chooses a free port for the TCP

server socket, and the port that it chose has to be published to the user via the parser interface.

5.119 io_tcp_parser_t Class Reference 381

Parameters

port The TCP port number that is currently used. In the range [MIN_TCP_PORT,

MAX_TCP_PORT], excluding 0.

Precondition

get_local_port() (p. 380) currently returns 0.

5.119.3.4 bool io_tcp_parser_t::get_server_port_open ( ) const [virtual]

Return the status of the server port as it is known to the parser.

Returns

false after initialization, or the value most recently set via

See also

set_server_port_open (p. 381).

5.119.3.5 void io_tcp_parser_t::set_server_port_open (

bool open )[virtual]

Inform the parser of the current status of the server socket.

Parameters

open Indicates wether the server socket has just been opened or closed.

Precondition

open may only have the value true if get_server_port_open() (p. 381) currently returns

false.

Postcondition

See also

get_server_port_open (p. 381) returns the value (p. 53) of open.

382 CONTENTS

5.119.3.6 bool io_tcp_parser_t::get_connected ( ) const [virtual]

Return the parser's knowledge concerning wether there currently exists an established sound

data TCP connection or not.

Returns

false after initialization, or the value most recently set via

See also

set_connected (p. 382).

5.119.3.7 void io_tcp_parser_t::set_connected (

bool connected )[virtual]

Inform the parser about the existance of a sound data connection.

Parameters

connected Indicates wether there currently is a connection or not.

Precondition

connected must not have the same value that is currently returned by

See also

get_connected (p. 382).

Postcondition

See also

get_connected (p. 382) returns the value (p. 53) of open.

5.119.3.8 void io_tcp_parser_t::set_new_peer (

unsigned short port,

const std::string & host )[virtual]

Set parser monitor variables peer_port and peer_address, and calls set_connected(true).

This method should be called when a new connection is established.

5.119 io_tcp_parser_t Class Reference 383

Parameters

port The TCP port number used by the peer.

host The Internet host where the peer is located.

Precondition

See also

get_connected (p. 382) currently returns false.

Postcondition

See also

get_connected (p. 382) returns true.

5.119.3.9 virtual void io_tcp_parser_t::debug (

const std::string & message )[inline],[virtual]

5.119.4 Member Data Documentation

5.119.4.1 MHAParser::string_t io_tcp_parser_t::local_address [private]

Lets the user set the local network interface to listen on.

5.119.4.2 MHAParser::int_t io_tcp_parser_t::local_port [private]

Lets the user choose the local tcp port to listen on.

5.119.4.3 MHAParser::int_mon_t io_tcp_parser_t::server_port_open [private]

Indicates wether the TCP server socket is currently open.

5.119.4.4 MHAParser::int_mon_t io_tcp_parser_t::connected [private]

Indicator if there currently is a sound data connection over TCP.

384 CONTENTS

5.119.4.5 MHAParser::string_mon_t io_tcp_parser_t::peer_address [private]

Display the ip address of the currently connected sound data client.

5.119.4.6 MHAParser::int_mon_t io_tcp_parser_t::peer_port [private]

Display the tcp port used by the current sound data client.

5.119.4.7 MHAParser::string_t io_tcp_parser_t::debug_ﬁlename [private]

ﬁlename to write debugging info to (if non-empty)

5.119.4.8 FILE∗io_tcp_parser_t::debug_ﬁle [private]

ﬁle handle to write debugging info to

The documentation for this class was generated from the following ﬁle:

•MHAIOTCP.cpp

5.120 io_tcp_sound_t Class Reference

Sound data handling of io tcp library.

Classes

• union ﬂoat_union

This union helps in conversion of ﬂoats from host byte order to network byte order and back

again.

Public Member Functions

•io_tcp_sound_t (int fragsize, ﬂoat samplerate)

Initialize sound data handling.

• virtual ∼io_tcp_sound_t ()

Do-nothing destructor.

• virtual void prepare (int num_inchannels, int num_outchannels)

Called during prepare, sets number of audio channels and allocates sound data storage.

• virtual void release ()

Called during release.

• virtual int chunkbytes_in () const

Number of bytes that constitute one input sound chunk.

• virtual std::string header () const

Create the tcp sound header lines.

• virtual mha_wave_t ∗ntoh (const std::string &data)

Copy data received from tcp into mha_wave_t (p. 496) structure.

• virtual std::string hton (const mha_wave_t ∗s_out)

Copy sound data from the output sound structure to a string.

5.120 io_tcp_sound_t Class Reference 385

Static Private Member Functions

• static void check_sound_data_type ()

Check if mha_real_t is a usable 32-bit ﬂoating point type.

Private Attributes

• int fragsize

Number of sound samples in each channel expected and returned from processing callback.

• ﬂoat samplerate

Sampling rate.

• int num_inchannels

Number of input channels.

• int num_outchannels

•MHASignal::waveform_t ∗s_in

Storage for input signal.

5.120.1 Detailed Description

Sound data handling of io tcp library.

5.120.2 Constructor & Destructor Documentation

5.120.2.1 io_tcp_sound_t::io_tcp_sound_t (

int fragsize,

ﬂoat samplerate )

Initialize sound data handling.

Checks sound data type by calling

See also

check_sound_data_type (p. 386).

Parameters

fragsize Number of sound samples in each channel expected and returned from

processing callback.

samplerate Number of samples per second in each channel.

386 CONTENTS

5.120.2.2 virtual io_tcp_sound_t::∼io_tcp_sound_t ( ) [inline],[virtual]

Do-nothing destructor.

5.120.3 Member Function Documentation

5.120.3.1 void io_tcp_sound_t::check_sound_data_type ( ) [static],[private]

Check if mha_real_t is a usable 32-bit ﬂoating point type.

Exceptions

MHA_Error (p. 445)if mha_real_t is not compatible to 32-bit ﬂoat.

5.120.3.2 void io_tcp_sound_t::prepare (

int num_inchannels,

int num_outchannels )[virtual]

Called during prepare, sets number of audio channels and allocates sound data storage.

Parameters

num_inchannels Number of input audio channels.

num_outchannels Number of output audio channels.

5.120.3.3 void io_tcp_sound_t::release (

void ) [virtual]

Called during release.

Deletes sound data storage.

5.120.3.4 int io_tcp_sound_t::chunkbytes_in ( ) const [virtual]

Number of bytes that constitute one input sound chunk.

Returns

Number of bytes to read from TCP connection before invoking signal processing.

5.120.3.5 std::string io_tcp_sound_t::header ( ) const [virtual]

Create the tcp sound header lines.

5.120 io_tcp_sound_t Class Reference 387

5.120.3.6 mha_wave_t ∗io_tcp_sound_t::ntoh (

const std::string & data )[virtual]

Copy data received from tcp into mha_wave_t (p. 496) structure.

Doing network-to-host byte order swapping in the process.

Parameters

data One chunk (

See also

chunkbytes_in (p. 386)) of sound data to process.

Returns

Pointer to the sound data storage.

5.120.3.7 std::string io_tcp_sound_t::hton (

const mha_wave_t ∗s_out )[virtual]

Copy sound data from the output sound structure to a string.

Doing host-to-network byte order swapping while at it.

Parameters

s_out Pointer to the storage of the sound to put out.

Returns

The sound data in network byte order.

5.120.4 Member Data Documentation

5.120.4.1 int io_tcp_sound_t::fragsize [private]

Number of sound samples in each channel expected and returned from processing callback.

5.120.4.2 ﬂoat io_tcp_sound_t::samplerate [private]

Sampling rate.

Number of samples per second in each channel.

388 CONTENTS

5.120.4.3 int io_tcp_sound_t::num_inchannels [private]

Number of input channels.

Number of channels expected from and returned by signal processing callback.

5.120.4.4 int io_tcp_sound_t::num_outchannels [private]

5.120.4.5 MHASignal::waveform_t∗io_tcp_sound_t::s_in [private]

Storage for input signal.

The documentation for this class was generated from the following ﬁle:

•MHAIOTCP.cpp

5.121 io_tcp_sound_t::ﬂoat_union Union Reference

This union helps in conversion of ﬂoats from host byte order to network byte order and back

again.

Public Attributes

• ﬂoat f

• unsigned int i

• char c[4]

5.121.1 Detailed Description

This union helps in conversion of ﬂoats from host byte order to network byte order and back

again.

5.121.2 Member Data Documentation

5.121.2.1 ﬂoat io_tcp_sound_t::ﬂoat_union::f

5.121.2.2 unsigned int io_tcp_sound_t::ﬂoat_union::i

5.121.2.3 char io_tcp_sound_t::ﬂoat_union::c[4]

The documentation for this union was generated from the following ﬁle:

•MHAIOTCP.cpp

5.122 io_tcp_t Class Reference 389

5.122 io_tcp_t Class Reference

The tcp sound io library.

Public Member Functions

•io_tcp_t (int fragsize, ﬂoat samplerate, IOProcessEvent_t proc_event, void ∗proc_←-

handle, IOStartedEvent_t start_event, void ∗start_handle, IOStoppedEvent_t stop_←-

event, void ∗stop_handle)

• void prepare (int num_inchannels, int num_outchannels)

Allocate server socket and start thread waiting for sound data exchange.

• void start ()

Call frameworks start callback if there is a sound data connection at the moment.

• void stop ()

Close the current connection if there is one.

• void release ()

Close the current connection and close the server socket.

• virtual void accept_loop ()

IO thread executes this method.

• virtual void connection_loop (MHA_TCP::Connection ∗c)

IO thread executes this method for each connection.

• virtual void parse (const char ∗cmd, char ∗retval, unsigned int len)

Parser interface.

• virtual ∼io_tcp_t ()

Private Attributes

•io_tcp_parser_t parser

•io_tcp_sound_t sound

•io_tcp_fwcb_t fwcb

•MHA_TCP::Server ∗server

•MHA_TCP::Thread ∗thread

•MHA_TCP::Async_Notify notify_start

•MHA_TCP::Async_Notify notify_stop

•MHA_TCP::Async_Notify notify_release

5.122.1 Detailed Description

The tcp sound io library.

390 CONTENTS

5.122.2 Constructor & Destructor Documentation

5.122.2.1 io_tcp_t::io_tcp_t (

int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

5.122.2.2 virtual io_tcp_t::∼io_tcp_t ( ) [inline],[virtual]

5.122.3 Member Function Documentation

5.122.3.1 void io_tcp_t::prepare (

int num_inchannels,

int num_outchannels )

Allocate server socket and start thread waiting for sound data exchange.

prepare opens the tcp server socket and starts the io thread that listens for audio data on the

tcp socket after doing some sanity checks

5.122.3.2 void io_tcp_t::start ( )

Call frameworks start callback if there is a sound data connection at the moment.

5.122.3.3 void io_tcp_t::stop ( )

Close the current connection if there is one.

stop IO thread

5.122.3.4 void io_tcp_t::release (

void )

Close the current connection and close the server socket.

Stop IO thread and close server socket.

5.122.3.5 void io_tcp_t::accept_loop ( ) [virtual]

IO thread executes this method.

5.122.3.6 void io_tcp_t::connection_loop (

MHA_TCP::Connection ∗c)[virtual]

IO thread executes this method for each connection.

5.123 latex_doc_t Class Reference 391

Parameters

cpointer to connection. connection_loop deletes connection before exiting.

5.122.3.7 virtual void io_tcp_t::parse (

const char ∗cmd,

char ∗retval,

unsigned int len )[inline],[virtual]

Parser interface.

5.122.4 Member Data Documentation

5.122.4.1 io_tcp_parser_t io_tcp_t::parser [private]

5.122.4.2 io_tcp_sound_t io_tcp_t::sound [private]

5.122.4.3 io_tcp_fwcb_t io_tcp_t::fwcb [private]

5.122.4.4 MHA_TCP::Server∗io_tcp_t::server [private]

5.122.4.5 MHA_TCP::Thread∗io_tcp_t::thread [private]

5.122.4.6 MHA_TCP::Async_Notify io_tcp_t::notify_start [private]

5.122.4.7 MHA_TCP::Async_Notify io_tcp_t::notify_stop [private]

5.122.4.8 MHA_TCP::Async_Notify io_tcp_t::notify_release [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOTCP.cpp

5.123 latex_doc_t Class Reference

Class to access the information stored in the plugin source code's MHAPLUGIN_DOCUME←-

NTATION macro.

392 CONTENTS

Public Member Functions

•latex_doc_t (const std::string &plugname, const std::string &plugin_macro)

Constructor loads the plugin into this process.

• std::string get_latex_doc ()

This method accesses the compiled-in contents of the MHAPLUGIN_DOCUMENTATION

macro and the exported interface functions of the loaded plugin to produce latex docuementa-

tion for the plugin.

• std::string get_main_category () const

• std::vector<std::string >get_categories () const

Private Member Functions

• std::string strdom (mha_domain_t d) const

• std::string get_ac (MHAKernel::algo_comm_class_t &ac, std::string txt) const

• std::string parsername (std::string s) const

• std::string get_parser_var (MHAParser::base_t ∗p, std::string name) const

• std::string get_parser_tab (MHAParser::base_t ∗p, const std::string &preﬁx, const std←-

::string &latex_macro) const

Private Attributes

• const std::string plugname

• const std::string latex_plugname

•MHAKernel::algo_comm_class_t ac

•PluginLoader::mhapluginloader_t loader

• const std::string plugin_macro

5.123.1 Detailed Description

Class to access the information stored in the plugin source code's MHAPLUGIN_DOCUME←-

NTATION macro.

5.123.2 Constructor & Destructor Documentation

5.123.2.1 latex_doc_t::latex_doc_t (

const std::string & plugname,

const std::string & plugin_macro )

Constructor loads the plugin into this process.

5.123 latex_doc_t Class Reference 393

Parameters

plugname Name of the MHA plugin to process

plugin_macro name of the LaTeX section macro that documents a single plugin (e.g.

"section", "subsection", "subsubsection", ...)

5.123.3 Member Function Documentation

5.123.3.1 std::string latex_doc_t::get_latex_doc ( )

This method accesses the compiled-in contents of the MHAPLUGIN_DOCUMENTATION

macro and the exported interface functions of the loaded plugin to produce latex docuementa-

tion for the plugin.

It tentatively prepares the plugin for processing and checks the AC variables registered by the

plugin.

Returns

the complete latex documentation for this plugin

5.123.3.2 std::string latex_doc_t::get_main_category ( ) const

Returns

the ﬁrst word of the categories string in the MHAPLUGIN_DOCUMENTATION macro

5.123.3.3 std::vector<std::string >latex_doc_t::get_categories ( ) const

Returns

a vector of all words in the categories string in the MHAPLUGIN_DOCUMENTATION

macro

394 CONTENTS

5.123.3.4 std::string latex_doc_t::strdom (

mha_domain_t d) const [private]

5.123.3.5 std::string latex_doc_t::get_ac (

MHAKernel::algo_comm_class_t &ac,

std::string txt ) const [private]

5.123.3.6 std::string latex_doc_t::parsername (

std::string s) const [private]

5.123.3.7 std::string latex_doc_t::get_parser_var (

MHAParser::base_t ∗p,

std::string name ) const [private]

5.123.3.8 std::string latex_doc_t::get_parser_tab (

MHAParser::base_t ∗p,

const std::string & preﬁx,

const std::string & latex_macro ) const [private]

5.123.4 Member Data Documentation

5.123.4.1 const std::string latex_doc_t::plugname [private]

5.123.4.2 const std::string latex_doc_t::latex_plugname [private]

5.123.4.3 MHAKernel::algo_comm_class_t latex_doc_t::ac [private]

5.123.4.4 PluginLoader::mhapluginloader_t latex_doc_t::loader [private]

5.123.4.5 const std::string latex_doc_t::plugin_macro [private]

The documentation for this class was generated from the following ﬁle:

•generatemhaplugindoc.cpp

5.124 lpc Class Reference 395

5.124 lpc Class Reference

Inheritance diagram for lpc:

lpc

MHAPlugin::plugin_t

< lpc_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< lpc_config >

Public Member Functions

•lpc (algo_comm_t &ac, const std::string &chain_name, const std::string &algo_name)

Constructs our plugin.

•∼lpc ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Private Member Functions

• void update_cfg ()

396 CONTENTS

Private Attributes

• std::string algo_name

•MHAParser::int_t lpc_order

•MHAParser::int_t lpc_buffer_size

•MHAParser::bool_t shift

•MHAParser::int_t comp_each_iter

•MHAParser::bool_t norm

•MHAEvents::patchbay_t<lpc >patchbay

Additional Inherited Members

5.124.1 Constructor & Destructor Documentation

5.124.1.1 lpc::lpc (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.124.1.2 lpc::∼lpc ( )

5.124.2 Member Function Documentation

5.124.2.1 mha_wave_t ∗lpc::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.124.2.2 void lpc::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<lpc_conﬁg >(p. 730).

5.124 lpc Class Reference 397

5.124.2.3 void lpc::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<lpc_conﬁg >(p. 731).

5.124.2.4 void lpc::update_cfg (

void ) [private]

5.124.3 Member Data Documentation

5.124.3.1 std::string lpc::algo_name [private]

5.124.3.2 MHAParser::int_t lpc::lpc_order [private]

5.124.3.3 MHAParser::int_t lpc::lpc_buffer_size [private]

5.124.3.4 MHAParser::bool_t lpc::shift [private]

5.124.3.5 MHAParser::int_t lpc::comp_each_iter [private]

5.124.3.6 MHAParser::bool_t lpc::norm [private]

5.124.3.7 MHAEvents::patchbay_t<lpc>lpc::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•lpc.h

•lpc.cpp

398 CONTENTS

5.125 lpc_bl_predictor Class Reference

Inheritance diagram for lpc_bl_predictor:

lpc_bl_predictor

MHAPlugin::plugin_t

< lpc_bl_predictor_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< lpc_bl_predictor_config >

Public Member Functions

•lpc_bl_predictor (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

Constructs our plugin.

•∼lpc_bl_predictor ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::int_t lpc_order

•MHAParser::string_t name_kappa

•MHAParser::string_t name_lpc_f

•MHAParser::string_t name_lpc_b

•MHAParser::string_t name_f

•MHAParser::string_t name_b

5.125 lpc_bl_predictor Class Reference 399

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<lpc_bl_predictor >patchbay

Additional Inherited Members

5.125.1 Constructor & Destructor Documentation

5.125.1.1 lpc_bl_predictor::lpc_bl_predictor (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

5.125.1.2 lpc_bl_predictor::∼lpc_bl_predictor ( )

5.125.2 Member Function Documentation

5.125.2.1 mha_wave_t ∗lpc_bl_predictor::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.125.2.2 void lpc_bl_predictor::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<lpc_bl_predictor_conﬁg >(p. 730).

400 CONTENTS

5.125.2.3 void lpc_bl_predictor::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<lpc_bl_predictor_conﬁg >(p. 731).

5.125.2.4 void lpc_bl_predictor::update_cfg (

void ) [private]

5.125.3 Member Data Documentation

5.125.3.1 MHAParser::int_t lpc_bl_predictor::lpc_order

5.125.3.2 MHAParser::string_t lpc_bl_predictor::name_kappa

5.125.3.3 MHAParser::string_t lpc_bl_predictor::name_lpc_f

5.125.3.4 MHAParser::string_t lpc_bl_predictor::name_lpc_b

5.125.3.5 MHAParser::string_t lpc_bl_predictor::name_f

5.125.3.6 MHAParser::string_t lpc_bl_predictor::name_b

5.125.3.7 MHAEvents::patchbay_t<lpc_bl_predictor>lpc_bl_predictor::patchbay

[private]

The documentation for this class was generated from the following ﬁles:

•lpc_bl_predictor.h

•lpc_bl_predictor.cpp

5.126 lpc_bl_predictor_conﬁg Class Reference

Public Member Functions

•lpc_bl_predictor_conﬁg (algo_comm_t &iac, const mhaconﬁg_t in_cfg, lpc_bl_←-

predictor ∗_lpc)

•∼lpc_bl_predictor_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

5.126 lpc_bl_predictor_conﬁg Class Reference 401

Private Attributes

•algo_comm_t ac

•MHA_AC::waveform_t f_est

•MHA_AC::waveform_t b_est

•MHASignal::waveform_t forward

•MHASignal::waveform_t backward

• int lpc_order

• std::string name_km

• std::string name_f

• std::string name_b

•mha_wave_t km

•mha_wave_t s_f

•mha_wave_t s_b

5.126.1 Constructor & Destructor Documentation

5.126.1.1 lpc_bl_predictor_conﬁg::lpc_bl_predictor_conﬁg (

algo_comm_t &iac,

const mhaconﬁg_t in_cfg,

lpc_bl_predictor ∗_lpc )

5.126.1.2 lpc_bl_predictor_conﬁg::∼lpc_bl_predictor_conﬁg ( )

5.126.2 Member Function Documentation

5.126.2.1 mha_wave_t ∗lpc_bl_predictor_conﬁg::process (

mha_wave_t ∗wave )

5.126.3 Member Data Documentation

5.126.3.1 algo_comm_t lpc_bl_predictor_conﬁg::ac [private]

5.126.3.2 MHA_AC::waveform_t lpc_bl_predictor_conﬁg::f_est [private]

5.126.3.3 MHA_AC::waveform_t lpc_bl_predictor_conﬁg::b_est [private]

5.126.3.4 MHASignal::waveform_t lpc_bl_predictor_conﬁg::forward [private]

5.126.3.5 MHASignal::waveform_t lpc_bl_predictor_conﬁg::backward [private]

5.126.3.6 int lpc_bl_predictor_conﬁg::lpc_order [private]

5.126.3.7 std::string lpc_bl_predictor_conﬁg::name_km [private]

402 CONTENTS

5.126.3.8 std::string lpc_bl_predictor_conﬁg::name_f [private]

5.126.3.9 std::string lpc_bl_predictor_conﬁg::name_b [private]

5.126.3.10 mha_wave_t lpc_bl_predictor_conﬁg::km [private]

5.126.3.11 mha_wave_t lpc_bl_predictor_conﬁg::s_f [private]

5.126.3.12 mha_wave_t lpc_bl_predictor_conﬁg::s_b [private]

The documentation for this class was generated from the following ﬁles:

•lpc_bl_predictor.h

•lpc_bl_predictor.cpp

5.127 lpc_burglattice Class Reference

Inheritance diagram for lpc_burglattice:

lpc_burglattice

MHAPlugin::plugin_t

< lpc_burglattice_config >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< lpc_burglattice_config >

5.127 lpc_burglattice Class Reference 403

Public Member Functions

•lpc_burglattice (algo_comm_t &ac, const std::string &chain_name, const std::string

&algo_name)

Constructs our plugin.

•∼lpc_burglattice ()

•mha_wave_t ∗process (mha_wave_t ∗)

Checks for the most recent conﬁguration and defers processing to it.

• void prepare (mhaconﬁg_t &)

Plugin preparation.

• void release (void)

Public Attributes

•MHAParser::int_t lpc_order

•MHAParser::string_t name_kappa

•MHAParser::string_t name_f

•MHAParser::string_t name_b

•MHAParser::ﬂoat_t lambda

Private Member Functions

• void update_cfg ()

Private Attributes

•MHAEvents::patchbay_t<lpc_burglattice >patchbay

Additional Inherited Members

5.127.1 Constructor & Destructor Documentation

5.127.1.1 lpc_burglattice::lpc_burglattice (

algo_comm_t &ac,

const std::string & chain_name,

const std::string & algo_name )

Constructs our plugin.

404 CONTENTS

5.127.1.2 lpc_burglattice::∼lpc_burglattice ( )

5.127.2 Member Function Documentation

5.127.2.1 mha_wave_t ∗lpc_burglattice::process (

mha_wave_t ∗signal )

Checks for the most recent conﬁguration and defers processing to it.

5.127.2.2 void lpc_burglattice::prepare (

mhaconﬁg_t &signal_info )[virtual]

Plugin preparation.

An opportunity to validate conﬁguration parameters before instantiating a conﬁguration.

Parameters

signal_info Structure containing a description of the form of the signal (domain, number of

channels, frames per block, sampling rate.

Implements MHAPlugin::plugin_t<lpc_burglattice_conﬁg >(p. 730).

5.127.2.3 void lpc_burglattice::release (

void ) [inline],[virtual]

Reimplemented from MHAPlugin::plugin_t<lpc_burglattice_conﬁg >(p. 731).

5.127.2.4 void lpc_burglattice::update_cfg (

void ) [private]

5.127.3 Member Data Documentation

5.127.3.1 MHAParser::int_t lpc_burglattice::lpc_order

5.127.3.2 MHAParser::string_t lpc_burglattice::name_kappa

5.127.3.3 MHAParser::string_t lpc_burglattice::name_f

5.127.3.4 MHAParser::string_t lpc_burglattice::name_b

5.127.3.5 MHAParser::ﬂoat_t lpc_burglattice::lambda

5.127.3.6 MHAEvents::patchbay_t<lpc_burglattice>lpc_burglattice::patchbay [private]

The documentation for this class was generated from the following ﬁles:

•lpc_burg-lattice.h

•lpc_burg-lattice.cpp

5.128 lpc_burglattice_conﬁg Class Reference 405

5.128 lpc_burglattice_conﬁg Class Reference

Public Member Functions

•lpc_burglattice_conﬁg (algo_comm_t &iac, const mhaconﬁg_t in_cfg, lpc_←-

burglattice ∗_lpc)

•∼lpc_burglattice_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

Private Attributes

•algo_comm_t ac

•MHASignal::waveform_t forward

•MHASignal::waveform_t backward

•MHASignal::waveform_t kappa

•MHA_AC::waveform_t kappa_block

•MHASignal::waveform_t dm

•MHASignal::waveform_t nm

•mha_real_t lambda

• int lpc_order

• std::string name_f

• std::string name_b

•mha_wave_t s_f

•mha_wave_t s_b

5.128.1 Constructor & Destructor Documentation

5.128.1.1 lpc_burglattice_conﬁg::lpc_burglattice_conﬁg (

algo_comm_t &iac,

const mhaconﬁg_t in_cfg,

lpc_burglattice ∗_lpc )

5.128.1.2 lpc_burglattice_conﬁg::∼lpc_burglattice_conﬁg ( )

5.128.2 Member Function Documentation

5.128.2.1 mha_wave_t ∗lpc_burglattice_conﬁg::process (

mha_wave_t ∗wave )

5.128.3 Member Data Documentation

5.128.3.1 algo_comm_t lpc_burglattice_conﬁg::ac [private]

406 CONTENTS

5.128.3.2 MHASignal::waveform_t lpc_burglattice_conﬁg::forward [private]

5.128.3.3 MHASignal::waveform_t lpc_burglattice_conﬁg::backward [private]

5.128.3.4 MHASignal::waveform_t lpc_burglattice_conﬁg::kappa [private]

5.128.3.5 MHA_AC::waveform_t lpc_burglattice_conﬁg::kappa_block [private]

5.128.3.6 MHASignal::waveform_t lpc_burglattice_conﬁg::dm [private]

5.128.3.7 MHASignal::waveform_t lpc_burglattice_conﬁg::nm [private]

5.128.3.8 mha_real_t lpc_burglattice_conﬁg::lambda [private]

5.128.3.9 int lpc_burglattice_conﬁg::lpc_order [private]

5.128.3.10 std::string lpc_burglattice_conﬁg::name_f [private]

5.128.3.11 std::string lpc_burglattice_conﬁg::name_b [private]

5.128.3.12 mha_wave_t lpc_burglattice_conﬁg::s_f [private]

5.128.3.13 mha_wave_t lpc_burglattice_conﬁg::s_b [private]

The documentation for this class was generated from the following ﬁles:

•lpc_burg-lattice.h

•lpc_burg-lattice.cpp

5.129 lpc_conﬁg Class Reference

Public Member Functions

•lpc_conﬁg (algo_comm_t &ac, const mhaconﬁg_t in_cfg, std::string &algo_name, un-

signed int _order, unsigned int _lpc_buffer_size, bool _shift, unsigned int _comp_each←-

_iter, bool _norm)

•∼lpc_conﬁg ()

•mha_wave_t ∗process (mha_wave_t ∗)

• void insert ()

5.129 lpc_conﬁg Class Reference 407

Private Attributes

• bool norm

• bool shift

• unsigned int comp_each_iter

• unsigned int order

• unsigned int lpc_buffer_size

• unsigned int N

• unsigned int comp_iter

•mha_wave_t sample

• std::vector<mha_real_t >R

• std::vector<mha_real_t >A

•MHASignal::ringbuffer_t inwave

•MHA_AC::waveform_t lpc_out

•MHA_AC::waveform_t corr_out

5.129.1 Constructor & Destructor Documentation

5.129.1.1 lpc_conﬁg::lpc_conﬁg (

algo_comm_t &ac,

const mhaconﬁg_t in_cfg,

std::string & algo_name,

unsigned int _order,

unsigned int _lpc_buffer_size,

bool _shift,

unsigned int _comp_each_iter,

bool _norm )

5.129.1.2 lpc_conﬁg::∼lpc_conﬁg ( )

5.129.2 Member Function Documentation

5.129.2.1 mha_wave_t ∗lpc_conﬁg::process (

mha_wave_t ∗wave )

5.129.2.2 void lpc_conﬁg::insert ( )

5.129.3 Member Data Documentation

5.129.3.1 bool lpc_conﬁg::norm [private]

5.129.3.2 bool lpc_conﬁg::shift [private]

5.129.3.3 unsigned int lpc_conﬁg::comp_each_iter [private]

408 CONTENTS

5.129.3.4 unsigned int lpc_conﬁg::order [private]

5.129.3.5 unsigned int lpc_conﬁg::lpc_buffer_size [private]

5.129.3.6 unsigned int lpc_conﬁg::N [private]

5.129.3.7 unsigned int lpc_conﬁg::comp_iter [private]

5.129.3.8 mha_wave_t lpc_conﬁg::sample [private]

5.129.3.9 std::vector<mha_real_t>lpc_conﬁg::R [private]

5.129.3.10 std::vector<mha_real_t>lpc_conﬁg::A [private]

5.129.3.11 MHASignal::ringbuffer_t lpc_conﬁg::inwave [private]

5.129.3.12 MHA_AC::waveform_t lpc_conﬁg::lpc_out [private]

5.129.3.13 MHA_AC::waveform_t lpc_conﬁg::corr_out [private]

The documentation for this class was generated from the following ﬁles:

•lpc.h

•lpc.cpp

5.130 matrixmixer::cfg_t Class Reference

Public Member Functions

•cfg_t (std::vector<std::vector<ﬂoat > > imixer, unsigned int ci, unsigned int co, un-

signed int fragsize, unsigned int nfft)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

Private Attributes

•MHASignal::waveform_t m

•MHASignal::waveform_t wout

•MHASignal::spectrum_t sout

5.130 matrixmixer::cfg_t Class Reference 409

5.130.1 Constructor & Destructor Documentation

5.130.1.1 cfg_t::cfg_t (

std::vector<std::vector<ﬂoat > > imixer,

unsigned int ci,

unsigned int co,

unsigned int fragsize,

unsigned int nfft )

5.130.2 Member Function Documentation

5.130.2.1 mha_wave_t ∗cfg_t::process (

mha_wave_t ∗s)

5.130.2.2 mha_spec_t ∗cfg_t::process (

mha_spec_t ∗s)

5.130.3 Member Data Documentation

5.130.3.1 MHASignal::waveform_t matrixmixer::cfg_t::m [private]

5.130.3.2 MHASignal::waveform_t matrixmixer::cfg_t::wout [private]

5.130.3.3 MHASignal::spectrum_t matrixmixer::cfg_t::sout [private]

The documentation for this class was generated from the following ﬁle:

•matrixmixer.cpp

410 CONTENTS

5.131 matrixmixer::matmix_t Class Reference

Inheritance diagram for matrixmixer::matmix_t:

matrixmixer::matmix_t

MHAPlugin::plugin_t

< cfg_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< cfg_t >

Public Member Functions

•matmix_t (const algo_comm_t &, const std::string &, const std::string &)

• void prepare (mhaconﬁg_t &)

•mha_wave_t ∗process (mha_wave_t ∗)

•mha_spec_t ∗process (mha_spec_t ∗)

Private Member Functions

• void update_m ()

Private Attributes

•MHAEvents::patchbay_t<matmix_t >patchbay

•MHAParser::mﬂoat_t mixer

• unsigned int ci

• unsigned int co

5.131 matrixmixer::matmix_t Class Reference 411

Additional Inherited Members

5.131.1 Constructor & Destructor Documentation

5.131.1.1 matrixmixer::matmix_t::matmix_t (

const algo_comm_t &iac,

const std::string & ,

const std::string & )

5.131.2 Member Function Documentation

5.131.2.1 void matrixmixer::matmix_t::prepare (

mhaconﬁg_t &tf )[virtual]

Implements MHAPlugin::plugin_t<cfg_t >(p. 730).

5.131.2.2 mha_wave_t ∗matrixmixer::matmix_t::process (

mha_wave_t ∗s)

5.131.2.3 mha_spec_t ∗matrixmixer::matmix_t::process (

mha_spec_t ∗s)

5.131.2.4 void matrixmixer::matmix_t::update_m (

void ) [private]

5.131.3 Member Data Documentation

5.131.3.1 MHAEvents::patchbay_t<matmix_t>matrixmixer::matmix_t::patchbay [private]

5.131.3.2 MHAParser::mﬂoat_t matrixmixer::matmix_t::mixer [private]

5.131.3.3 unsigned int matrixmixer::matmix_t::ci [private]

5.131.3.4 unsigned int matrixmixer::matmix_t::co [private]

The documentation for this class was generated from the following ﬁle:

•matrixmixer.cpp

412 CONTENTS

5.132 MHA_AC::ac2matrix_helper_t Class Reference

Inheritance diagram for MHA_AC::ac2matrix_helper_t:

MHA_AC::ac2matrix_helper_t

MHA_AC::ac2matrix_t

Public Member Functions

•ac2matrix_helper_t (algo_comm_t, const std::string &)

• void getvar ()

Public Attributes

•algo_comm_t ac

• std::string name

• std::string username

•MHASignal::uint_vector_t size

• bool is_complex

Protected Attributes

•comm_var_t acvar

5.132.1 Constructor & Destructor Documentation

5.132.1.1 MHA_AC::ac2matrix_helper_t::ac2matrix_helper_t (

algo_comm_t iac,

const std::string & iname )

5.132.2 Member Function Documentation

5.132.2.1 void MHA_AC::ac2matrix_helper_t::getvar ( )

5.133 MHA_AC::ac2matrix_t Class Reference 413

5.132.3 Member Data Documentation

5.132.3.1 algo_comm_t MHA_AC::ac2matrix_helper_t::ac

5.132.3.2 std::string MHA_AC::ac2matrix_helper_t::name

5.132.3.3 std::string MHA_AC::ac2matrix_helper_t::username

5.132.3.4 MHASignal::uint_vector_t MHA_AC::ac2matrix_helper_t::size

5.132.3.5 bool MHA_AC::ac2matrix_helper_t::is_complex

5.132.3.6 comm_var_t MHA_AC::ac2matrix_helper_t::acvar [protected]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.133 MHA_AC::ac2matrix_t Class Reference

Copy AC variable to a matrix.

Inheritance diagram for MHA_AC::ac2matrix_t:

MHA_AC::ac2matrix_t

MHA_AC::ac2matrix_helper_t MHASignal::matrix_t

MHASignal::uint_vector_t

414 CONTENTS

Public Member Functions

•ac2matrix_t (algo_comm_t ac, const std::string &name)

Constructor.

• void update ()

Update contents of the matrix from the AC space.

• const std::string & getname () const

Return name of AC variable/matrix.

• const std::string & getusername () const

Return user speciﬁed name of AC variable/matrix.

• void insert (algo_comm_t ac)

Insert matrix into an AC space (other than source AC space)

Additional Inherited Members

5.133.1 Detailed Description

Copy AC variable to a matrix.

This class constructs a matrix of same size as an AC variable and can copy the AC variable to

itself. The update() (p. 414) function is real-time safe.

5.133.2 Constructor & Destructor Documentation

5.133.2.1 MHA_AC::ac2matrix_t::ac2matrix_t (

algo_comm_t ac,

const std::string & name )

Constructor.

Parameters

ac AC handle

name Name of AC variable to be copied

5.133.3 Member Function Documentation

5.133.3.1 void MHA_AC::ac2matrix_t::update ( )

Update contents of the matrix from the AC space.

This function is real-time safe. The copy operation performance is of the order of the number

of elements in the matrix.

5.134 MHA_AC::acspace2matrix_t Class Reference 415

5.133.3.2 const std::string& MHA_AC::ac2matrix_t::getname ( ) const [inline]

Return name of AC variable/matrix.

5.133.3.3 const std::string& MHA_AC::ac2matrix_t::getusername ( ) const [inline]

Return user speciﬁed name of AC variable/matrix.

5.133.3.4 void MHA_AC::ac2matrix_t::insert (

algo_comm_t ac )

Insert matrix into an AC space (other than source AC space)

Parameters

ac AC space handle to insert data

Note

The AC variable data buffer points to the data of the matrix. Modiﬁcations of the AC

variable directly modify the data of the matrix; after deletion of the matrix, the data buffer

is invalid.

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.134 MHA_AC::acspace2matrix_t Class Reference

Copy all or a subset of all numeric AC variables into an array of matrixes.

Public Member Functions

•acspace2matrix_t (algo_comm_t ac, const std::vector<std::string >&names)

Constructor.

•acspace2matrix_t (const MHA_AC::acspace2matrix_t &src)

Constructor with initialization from an instance.

•∼acspace2matrix_t ()

•MHA_AC::acspace2matrix_t &operator= (const MHA_AC::acspace2matrix_t &src)

Copy all contents (deep copy).

•MHA_AC::ac2matrix_t &operator[ ] (unsigned int k)

Access operator.

416 CONTENTS

• const MHA_AC::ac2matrix_t &operator[ ] (unsigned int k) const

Constant access operator.

• void update ()

Update function.

• unsigned int size () const

Number of matrixes in AC space.

• unsigned int frame () const

Actual frame number.

• void insert (algo_comm_t ac)

Insert AC space copy into an AC space (other than source AC space)

Private Attributes

• unsigned int len

•MHA_AC::ac2matrix_t ∗∗ data

• unsigned int frameno

5.134.1 Detailed Description

Copy all or a subset of all numeric AC variables into an array of matrixes.

5.134.2 Constructor & Destructor Documentation

5.134.2.1 MHA_AC::acspace2matrix_t::acspace2matrix_t (

algo_comm_t ac,

const std::vector<std::string >&names )

Constructor.

Scan all given AC variables and allocate corresponding matrixes.

Parameters

ac AC handle.

names Names of AC variables, or empty for all.

5.134.2.2 MHA_AC::acspace2matrix_t::acspace2matrix_t (

const MHA_AC::acspace2matrix_t &src )

Constructor with initialization from an instance.

5.134 MHA_AC::acspace2matrix_t Class Reference 417

Parameters

src Instance to be copied.

5.134.2.3 MHA_AC::acspace2matrix_t::∼acspace2matrix_t ( )

5.134.3 Member Function Documentation

5.134.3.1 MHA_AC::acspace2matrix_t & MHA_AC::acspace2matrix_t::operator= (

const MHA_AC::acspace2matrix_t &src )

Copy all contents (deep copy).

Parameters

src Array of matrixes to be copied.

5.134.3.2 MHA_AC::ac2matrix_t& MHA_AC::acspace2matrix_t::operator[ ] (

unsigned int k)[inline]

Access operator.

Parameters

kindex into array; should not exceed size() (p. 418)-1.

Return values

Reference to matrix.

5.134.3.3 const MHA_AC::ac2matrix_t& MHA_AC::acspace2matrix_t::operator[ ] (

unsigned int k) const [inline]

Constant access operator.

Parameters

kindex into array; should not exceed size() (p. 418)-1.

Return values

Constant reference to matrix.

418 CONTENTS

5.134.3.4 void MHA_AC::acspace2matrix_t::update ( ) [inline]

Update function.

This function updates all matrixes from their corresponding AC variables. It can be called from

the MHA Framework prepare function or in the processing callback.

5.134.3.5 unsigned int MHA_AC::acspace2matrix_t::size ( ) const [inline]

Number of matrixes in AC space.

5.134.3.6 unsigned int MHA_AC::acspace2matrix_t::frame ( ) const [inline]

Actual frame number.

5.134.3.7 void MHA_AC::acspace2matrix_t::insert (

algo_comm_t ac )

Insert AC space copy into an AC space (other than source AC space)

Parameters

ac AC space handle to insert data

5.134.4 Member Data Documentation

5.134.4.1 unsigned int MHA_AC::acspace2matrix_t::len [private]

5.134.4.2 MHA_AC::ac2matrix_t∗∗ MHA_AC::acspace2matrix_t::data [private]

5.134.4.3 unsigned int MHA_AC::acspace2matrix_t::frameno [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.135 MHA_AC::double_t Class Reference

Insert a double precision ﬂoating point variable into the AC space.

5.135 MHA_AC::double_t Class Reference 419

Public Member Functions

•double_t (algo_comm_t, std::string, double=0)

•∼double_t ()

Public Attributes

• double data

Floating point value variable.

Private Attributes

•algo_comm_t ac

5.135.1 Detailed Description

Insert a double precision ﬂoating point variable into the AC space.

The variable is automatically removed on destruction.

5.135.2 Constructor & Destructor Documentation

5.135.2.1 MHA_AC::double_t::double_t (

algo_comm_t iac,

std::string n,

double v = 0)

5.135.2.2 MHA_AC::double_t::∼double_t ( )

5.135.3 Member Data Documentation

5.135.3.1 double MHA_AC::double_t::data

Floating point value variable.

5.135.3.2 algo_comm_t MHA_AC::double_t::ac [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

420 CONTENTS

5.136 MHA_AC::ﬂoat_t Class Reference

Insert a ﬂoat point variable into the AC space.

Public Member Functions

•ﬂoat_t (algo_comm_t, std::string, ﬂoat=0)

Constructor.

•∼ﬂoat_t ()

Public Attributes

• ﬂoat data

Floating point value variable.

Private Attributes

•algo_comm_t ac

5.136.1 Detailed Description

Insert a ﬂoat point variable into the AC space.

The variable is automatically removed on destruction.

5.136.2 Constructor & Destructor Documentation

5.136.2.1 MHA_AC::ﬂoat_t::ﬂoat_t (

algo_comm_t iac,

std::string n,

ﬂoat v = 0)

Constructor.

5.136.2.2 MHA_AC::ﬂoat_t::∼ﬂoat_t ( )

5.136.3 Member Data Documentation

5.136.3.1 ﬂoat MHA_AC::ﬂoat_t::data

Floating point value variable.

5.137 MHA_AC::int_t Class Reference 421

5.136.3.2 algo_comm_t MHA_AC::ﬂoat_t::ac [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.137 MHA_AC::int_t Class Reference

Insert a integer variable into the AC space.

Public Member Functions

•int_t (algo_comm_t, std::string, int=0)

•∼int_t ()

Public Attributes

• int data

Integer value variable.

Private Attributes

•algo_comm_t ac

5.137.1 Detailed Description

Insert a integer variable into the AC space.

The variable is automatically removed on destruction.

5.137.2 Constructor & Destructor Documentation

5.137.2.1 MHA_AC::int_t::int_t (

algo_comm_t iac,

std::string n,

int v = 0)

5.137.2.2 MHA_AC::int_t::∼int_t ( )

5.137.3 Member Data Documentation

5.137.3.1 int MHA_AC::int_t::data

Integer value variable.

422 CONTENTS

5.137.3.2 algo_comm_t MHA_AC::int_t::ac [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.138 MHA_AC::spectrum_t Class Reference

Insert a MHASignal::spectrum_t (p. 800) class into the AC space.

Inheritance diagram for MHA_AC::spectrum_t:

MHA_AC::spectrum_t

MHAMultiSrc::spectrum_t wave2spec_t

MHASignal::spectrum_t

mha_spec_t

Public Member Functions

•spectrum_t (algo_comm_t ac, std::string name, unsigned int bins, unsigned int chan-

nels, bool insert_now)

Create the AC variable.

•∼spectrum_t ()

• void insert ()

Insert AC variable into AC space.

Protected Attributes

•algo_comm_t ac

• std::string name

5.138 MHA_AC::spectrum_t Class Reference 423

Additional Inherited Members

5.138.1 Detailed Description

Insert a MHASignal::spectrum_t (p. 800) class into the AC space.

The variable is automatically removed on destruction.

5.138.2 Constructor & Destructor Documentation

5.138.2.1 MHA_AC::spectrum_t::spectrum_t (

algo_comm_t ac,

std::string name,

unsigned int bins,

unsigned int channels,

bool insert_now )

Create the AC variable.

Parameters

ac AC handle

name Name of variable in AC space

bins Number of FFT bins in the waveform_t (p. 425) class

channels Number of audio channels in the waveform_t (p. 425) class

insert_now Insert implicitely in the constructor (true) or explicitely in the insert() (p. 423)

function (false)

5.138.2.2 MHA_AC::spectrum_t::∼spectrum_t (

void ) [virtual]

Reimplemented from MHASignal::spectrum_t (p. 802).

5.138.3 Member Function Documentation

5.138.3.1 void MHA_AC::spectrum_t::insert ( )

Insert AC variable into AC space.

424 CONTENTS

5.138.4 Member Data Documentation

5.138.4.1 algo_comm_t MHA_AC::spectrum_t::ac [protected]

5.138.4.2 std::string MHA_AC::spectrum_t::name [protected]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.139 MHA_AC::stat_t Class Reference

Inheritance diagram for MHA_AC::stat_t:

MHA_AC::stat_t

MHASignal::stat_t

Public Member Functions

•stat_t (algo_comm_t ac, const std::string &name, const unsigned int &frames, const

unsigned int &channels, bool insert_now)

• void update ()

• void insert ()

Private Attributes

•MHA_AC::waveform_t mean

•MHA_AC::waveform_t std

5.140 MHA_AC::waveform_t Class Reference 425

5.139.1 Constructor & Destructor Documentation

5.139.1.1 MHA_AC::stat_t::stat_t (

algo_comm_t ac,

const std::string & name,

const unsigned int & frames,

const unsigned int & channels,

bool insert_now )

5.139.2 Member Function Documentation

5.139.2.1 void MHA_AC::stat_t::update ( )

5.139.2.2 void MHA_AC::stat_t::insert ( )

5.139.3 Member Data Documentation

5.139.3.1 MHA_AC::waveform_t MHA_AC::stat_t::mean [private]

5.139.3.2 MHA_AC::waveform_t MHA_AC::stat_t::std [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.140 MHA_AC::waveform_t Class Reference

Insert a MHASignal::waveform_t (p. 812) class into the AC space.

Inheritance diagram for MHA_AC::waveform_t:

MHA_AC::waveform_t

MHAMultiSrc::waveform_t mon_t

MHASignal::waveform_t

mha_wave_t

426 CONTENTS

Public Member Functions

•waveform_t (algo_comm_t ac, std::string name, unsigned int frames, unsigned int

channels, bool insert_now)

Create the AC variable.

•∼waveform_t ()

• void insert ()

Insert AC variable into AC space.

Protected Attributes

•algo_comm_t ac

• std::string name

Additional Inherited Members

5.140.1 Detailed Description

Insert a MHASignal::waveform_t (p. 812) class into the AC space.

The variable is automatically removed on destruction.

5.140.2 Constructor & Destructor Documentation

5.140.2.1 MHA_AC::waveform_t::waveform_t (

algo_comm_t ac,

std::string name,

unsigned int frames,

unsigned int channels,

bool insert_now )

Create the AC variable.

Parameters

ac AC handle

name Name of variable in AC space

frames Number of frames in the waveform_t (p. 425) class

channels Number of audio channels in the waveform_t (p. 425) class

insert_now Insert implicitely in the constructor (true) or explicitely in the insert() (p. 427)

function (false)

5.141 mha_audio_descriptor_t Struct Reference 427

5.140.2.2 MHA_AC::waveform_t::∼waveform_t (

void ) [virtual]

Reimplemented from MHASignal::waveform_t (p. 815).

5.140.3 Member Function Documentation

5.140.3.1 void MHA_AC::waveform_t::insert ( )

Insert AC variable into AC space.

5.140.4 Member Data Documentation

5.140.4.1 algo_comm_t MHA_AC::waveform_t::ac [protected]

5.140.4.2 std::string MHA_AC::waveform_t::name [protected]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.h

•mha_algo_comm.cpp

5.141 mha_audio_descriptor_t Struct Reference

Description of an audio fragment (planned as a replacement of mhaconﬁg_t (p. 504)).

Public Attributes

• unsigned int n_samples

Number of samples.

• unsigned int n_channels

Number of audio channels.

• unsigned int n_freqs

Number of frequency bands.

• unsigned int is_complex

Flag about sample type.

•mha_real_t dt

Time distance between samples (only equidistant samples allowed)

•mha_real_t ∗cf

Center frequencies of frequency bands.

•mha_real_t ∗chdir

Hint on source direction of channel, values below zero is left, values above zero is right, zero

means unknown.

428 CONTENTS

5.141.1 Detailed Description

Description of an audio fragment (planned as a replacement of mhaconﬁg_t (p. 504)).

5.141.2 Member Data Documentation

5.141.2.1 unsigned int mha_audio_descriptor_t::n_samples

Number of samples.

5.141.2.2 unsigned int mha_audio_descriptor_t::n_channels

Number of audio channels.

5.141.2.3 unsigned int mha_audio_descriptor_t::n_freqs

Number of frequency bands.

5.141.2.4 unsigned int mha_audio_descriptor_t::is_complex

Flag about sample type.

5.141.2.5 mha_real_t mha_audio_descriptor_t::dt

Time distance between samples (only equidistant samples allowed)

5.141.2.6 mha_real_t∗mha_audio_descriptor_t::cf

Center frequencies of frequency bands.

5.141.2.7 mha_real_t∗mha_audio_descriptor_t::chdir

Hint on source direction of channel, values below zero is left, values above zero is right, zero

means unknown.

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.142 mha_audio_t Struct Reference

An audio fragment in the openMHA (planned as a replacement of mha_wave_t (p. 496) and

mha_spec_t (p. 465)).

5.143 mha_channel_info_t Struct Reference 429

Public Attributes

•mha_audio_descriptor_t descriptor

Dimension and description of the data.

•mha_real_t ∗rdata

Data pointer if ﬂag mha_audio_descriptor_t::is_complex (p. 428) is unset.

•mha_complex_t ∗cdata

Data pointer if ﬂag mha_audio_descriptor_t::is_complex (p. 428) is set.

5.142.1 Detailed Description

An audio fragment in the openMHA (planned as a replacement of mha_wave_t (p. 496) and

mha_spec_t (p. 465)).

The data alignment is (t0, c0, f0),(t0, c0, f1),...,(t0, c0, ffreqs),(t0, c1, f0), . . .. This allows a di-

rect cast of the current mha_wave_t (p. 496) and mha_spec_t (p. 465) data pointers into cor-

responding mha_audio_t (p. 428) objects.

5.142.2 Member Data Documentation

5.142.2.1 mha_audio_descriptor_t mha_audio_t::descriptor

Dimension and description of the data.

5.142.2.2 mha_real_t∗mha_audio_t::rdata

Data pointer if ﬂag mha_audio_descriptor_t::is_complex (p. 428) is unset.

5.142.2.3 mha_complex_t∗mha_audio_t::cdata

Data pointer if ﬂag mha_audio_descriptor_t::is_complex (p. 428) is set.

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.143 mha_channel_info_t Struct Reference

Channel information structure.

430 CONTENTS

Public Attributes

• int id

channel id

• char idstr [32]

channel id

• unsigned int side

side (left/right)

•mha_direction_t dir

source direction

•mha_real_t peaklevel

Peak level corresponds to this SPL (dB) level.

5.143.1 Detailed Description

Channel information structure.

5.143.2 Member Data Documentation

5.143.2.1 int mha_channel_info_t::id

channel id

5.143.2.2 char mha_channel_info_t::idstr[32]

channel id

5.143.2.3 unsigned int mha_channel_info_t::side

side (left/right)

5.143.2.4 mha_direction_t mha_channel_info_t::dir

source direction

5.143.2.5 mha_real_t mha_channel_info_t::peaklevel

Peak level corresponds to this SPL (dB) level.

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.144 mha_complex_t Struct Reference 431

5.144 mha_complex_t Struct Reference

Type for complex ﬂoating point values.

Public Attributes

•mha_real_t re

Real part.

•mha_real_t im

Imaginary part.

5.144.1 Detailed Description

Type for complex ﬂoating point values.

5.144.2 Member Data Documentation

5.144.2.1 mha_real_t mha_complex_t::re

Real part.

5.144.2.2 mha_real_t mha_complex_t::im

Imaginary part.

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.145 mha_complex_test_array_t Struct Reference

Several places in MHA rely on the fact that you can cast an array of mha_complex_t (p. 431)

c[] to an array of mha_real_t r[] with r[0] == c[0].re r[1] == c[0].im r[2] == c[1].re ...

Public Attributes

•mha_complex_t c [2]

432 CONTENTS

5.145.1 Detailed Description

Several places in MHA rely on the fact that you can cast an array of mha_complex_t (p. 431)

c[] to an array of mha_real_t r[] with r[0] == c[0].re r[1] == c[0].im r[2] == c[1].re ...

Check these expectations in static asserts.

5.145.2 Member Data Documentation

5.145.2.1 mha_complex_t mha_complex_test_array_t::c[2]

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.146 mha_dblbuf_t<FIFO >Class Template Reference

The doublebuffer adapts blocksizes between an outer process, which provides input data and

takes output data, and an inner process, which processes the input signal and generates output

data using a different block size than the outer process.

Public Types

• typedef FIFO::value_type value_type

The datatype exchanged by the FIFO and this doublebuffer.

Public Member Functions

• virtual unsigned get_inner_size () const

• virtual unsigned get_outer_size () const

• virtual unsigned get_delay () const

• virtual unsigned get_ﬁfo_size () const

• virtual unsigned get_input_channels () const

• virtual unsigned get_output_channels () const

• virtual unsigned get_input_ﬁfo_ﬁll_count () const

• virtual unsigned get_output_ﬁfo_ﬁll_count () const

• virtual unsigned get_input_ﬁfo_space () const

• virtual unsigned get_output_ﬁfo_space () const

• virtual MHA_Error ∗get_inner_error () const

• virtual void provoke_inner_error (const MHA_Error &)

• virtual void provoke_outer_error (const MHA_Error &)

5.146 mha_dblbuf_t<FIFO >Class Template Reference 433

•mha_dblbuf_t (unsigned outer_size, unsigned inner_size, unsigned delay, unsigned

input_channels, unsigned output_channels, const value_type &delay_data)

Constructor creates FIFOs with speciﬁed delay.

• virtual ∼mha_dblbuf_t ()

• virtual void process (const value_type ∗input_signal, value_type ∗output_signal, un-

signed count)

The outer process has to call this method to propagate the input signal to the inner process,

and receives back the output signal.

• virtual void input (value_type ∗input_signal)

The inner process has to call this method to receive its input signal.

• virtual void output (const value_type ∗output_signal)

The outer process has to call this method to deliver its output signal.

Private Attributes

• unsigned outer_size

The block size used by the outer process.

• unsigned inner_size

The block size used by the inner process.

• unsigned delay

The delay introduced by bidirectional buffer size adaptation.

• unsigned ﬁfo_size

The size of each of the FIFOs.

• unsigned input_channels

The number of input channels.

• unsigned output_channels

The number of output channels.

• FIFO input_ﬁfo

The FIFO for transporting the input signal from the outer process to the inner process.

• FIFO output_ﬁfo

The FIFO for transporting the output signal from the inner process to the outer process.

•MHA_Error ∗inner_error

Owned copy of exception to be thrown in inner thread.

•MHA_Error ∗outer_error

Owned copy of exception to be thrown in outer thread.

5.146.1 Detailed Description

template<class FIFO>

class mha_dblbuf_t<FIFO >

The doublebuffer adapts blocksizes between an outer process, which provides input data and

takes output data, and an inner process, which processes the input signal and generates output

data using a different block size than the outer process.

This class introduces the channels concept. Input and output may have different channel

counts.

434 CONTENTS

5.146.2 Member Typedef Documentation

5.146.2.1 template<class FIFO >typedef FIFO::value_type mha_dblbuf_t<FIFO >::value_type

The datatype exchanged by the FIFO and this doublebuffer.

5.146.3 Constructor & Destructor Documentation

5.146.3.1 template<class FIFO >mha_dblbuf_t<FIFO >::mha_dblbuf_t (

unsigned outer_size,

unsigned inner_size,

unsigned delay,

unsigned input_channels,

unsigned output_channels,

const value_type &delay_data )

Constructor creates FIFOs with speciﬁed delay.

Warning

The doublebuffer may block or raise an exception if the delay is too small. To avoid this,

the delay should be

delay >= (inner_size −gcd(inner_size, outer_size))

Parameters

outer_size The block size used by the outer process.

inner_size The block size used by the inner process.

delay The total delay

input_channels Number of input channels

output_channels Number of output channels

delay_data The delay consists of copies of this value.

5.146.3.2 template<class FIFO >mha_dblbuf_t<FIFO >::∼mha_dblbuf_t ( ) [virtual]

5.146.4 Member Function Documentation

5.146.4.1 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_inner_size ( ) const

[inline],[virtual]

5.146 mha_dblbuf_t<FIFO >Class Template Reference 435

5.146.4.2 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_outer_size ( ) const

[inline],[virtual]

5.146.4.3 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_delay ( ) const

[inline],[virtual]

5.146.4.4 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_ﬁfo_size ( ) const

[inline],[virtual]

5.146.4.5 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_input_channels ( )

const [inline],[virtual]

5.146.4.6 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_output_channels ( )

const [inline],[virtual]

5.146.4.7 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_input_ﬁfo_ﬁll_count (

) const [inline],[virtual]

5.146.4.8 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_output_ﬁfo_ﬁll_count

( ) const [inline],[virtual]

5.146.4.9 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_input_ﬁfo_space ( )

const [inline],[virtual]

5.146.4.10 template<class FIFO >virtual unsigned mha_dblbuf_t<FIFO >::get_output_ﬁfo_space (

) const [inline],[virtual]

5.146.4.11 template<class FIFO >virtual MHA_Error∗mha_dblbuf_t<FIFO >::get_inner_error (

) const [inline],[virtual]

5.146.4.12 template<class FIFO >void mha_dblbuf_t<FIFO >::provoke_inner_error (

const MHA_Error &error )[virtual]

5.146.4.13 template<class FIFO >void mha_dblbuf_t<FIFO >::provoke_outer_error (

const MHA_Error &error )[virtual]

5.146.4.14 template<class FIFO >void mha_dblbuf_t<FIFO >::process (

const value_type ∗input_signal,

value_type ∗output_signal,

unsigned count )[virtual]

The outer process has to call this method to propagate the input signal to the inner process,

and receives back the output signal.

436 CONTENTS

Parameters

input_signal Pointer to the input signal array.

output_signal Pointer to the output signal array.

count The number of data instances provided and expected, lower or equal to

inner_size given to constructor.

Exceptions

MHA_Error (p. 445)When count is >outer_size as given to constructor or the underlying

ﬁfo implementation detects an error.

5.146.4.15 template<class FIFO >void mha_dblbuf_t<FIFO >::input (

value_type ∗input_signal )[virtual]

The inner process has to call this method to receive its input signal.

Parameters

input_signal Array where the doublebuffer can store the signal.

Exceptions

MHA_Error (p. 445)When the underlying ﬁfo implementation detects an error.

5.146.4.16 template<class FIFO >void mha_dblbuf_t<FIFO >::output (

const value_type ∗output_signal )[virtual]

The outer process has to call this method to deliver its output signal.

Parameters

output_signal Array from which doublebuffer reads outputsignal.

Exceptions

MHA_Error (p. 445)When the underlying ﬁfo implementation detects an error.

5.146.5 Member Data Documentation

5.146 mha_dblbuf_t<FIFO >Class Template Reference 437

5.146.5.1 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::outer_size [private]

The block size used by the outer process.

5.146.5.2 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::inner_size [private]

The block size used by the inner process.

5.146.5.3 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::delay [private]

The delay introduced by bidirectional buffer size adaptation.

5.146.5.4 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::ﬁfo_size [private]

The size of each of the FIFOs.

5.146.5.5 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::input_channels [private]

The number of input channels.

5.146.5.6 template<class FIFO >unsigned mha_dblbuf_t<FIFO >::output_channels [private]

The number of output channels.

5.146.5.7 template<class FIFO >FIFO mha_dblbuf_t<FIFO >::input_ﬁfo [private]

The FIFO for transporting the input signal from the outer process to the inner process.

5.146.5.8 template<class FIFO >FIFO mha_dblbuf_t<FIFO >::output_ﬁfo [private]

The FIFO for transporting the output signal from the inner process to the outer process.

5.146.5.9 template<class FIFO >MHA_Error∗mha_dblbuf_t<FIFO >::inner_error [private]

Owned copy of exception to be thrown in inner thread.

5.146.5.10 template<class FIFO >MHA_Error∗mha_dblbuf_t<FIFO >::outer_error

[private]

Owned copy of exception to be thrown in outer thread.

The documentation for this class was generated from the following ﬁles:

•mha_ﬁfo.h

•mha_ﬁfo.cpp

438 CONTENTS

5.147 mha_direction_t Struct Reference

Channel source direction structure.

Public Attributes

•mha_real_t azimuth

azimuth in radiants

•mha_real_t elevation

elevation in radiants

•mha_real_t distance

distance in meters

5.147.1 Detailed Description

Channel source direction structure.

5.147.2 Member Data Documentation

5.147.2.1 mha_real_t mha_direction_t::azimuth

azimuth in radiants

5.147.2.2 mha_real_t mha_direction_t::elevation

elevation in radiants

5.147.2.3 mha_real_t mha_direction_t::distance

distance in meters

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.148 mha_drifter_ﬁfo_t<T>Class Template Reference 439

5.148 mha_drifter_ﬁfo_t<T>Class Template Reference

A FIFO class for blocksize adaptation without Synchronization.

Inheritance diagram for mha_drifter_ﬁfo_t<T>:

mha_drifter_fifo_t< T >

mha_fifo_t< T >

Public Member Functions

• virtual void write (const T ∗data, unsigned count)

write data to ﬁfo

• virtual void read (T ∗buf, unsigned count)

Read data from ﬁfo.

• virtual unsigned get_ﬁll_count () const

Return ﬁll_count, adding mha_drifter_ﬁfo_t<T>::startup_zeros (p. 445) to the number of

samples actually in the ﬁfo's buffer.

• virtual unsigned get_available_space () const

Return available space, subtracting number of mha_drifter_ﬁfo_t<T>::startup_zeros

(p. 445) from the available_space actually present in the ﬁfo's buffer.

• virtual unsigned get_des_ﬁll_count () const

The desired ﬁll count of this ﬁfo.

• virtual unsigned get_min_ﬁll_count () const

The minimum ﬁll count of this ﬁfo.

• virtual void stop ()

Called by mha_drifter_ﬁfo_t<T>::read (p. 441) or mha_drifter_ﬁfo_t<T>::write (p. 441)

when their xrun in succession counter exceeds its limit.

• virtual void starting ()

Called by mha_drifter_ﬁfo_t<T>::read (p. 441) or mha_drifter_ﬁfo_t<T>::write (p. 441)

when the respective ﬂag (mha_drifter_ﬁfo_t<T>::reader_started (p. 443) or mha_drifter←-

_ﬁfo_t<T>::writer_started (p. 443)) is about to be toggled from false to true.

•mha_drifter_ﬁfo_t (unsigned min_ﬁll_count, unsigned desired_ﬁll_count, unsigned

max_ﬁll_count)

Create drifter FIFO.

•mha_drifter_ﬁfo_t (unsigned min_ﬁll_count, unsigned desired_ﬁll_count, unsigned

max_ﬁll_count, const T &t)

Create drifter FIFO where all (initially unused) copies of T are initialized as copies of t.

440 CONTENTS

Private Attributes

• const unsigned minimum_ﬁll_count

The minimum ﬁll count of this ﬁfo.

• const unsigned desired_ﬁll_count

The desired ﬁll count of the ﬁfo.

• bool writer_started

Flag set to true when write is called the ﬁrst time.

• bool reader_started

Flag set to true when read is called for the ﬁrst time.

• unsigned writer_xruns_total

The number of xruns seen by the writer since object instantiation.

• unsigned reader_xruns_total

The number of xruns seen by the reader since object instantiation.

• unsigned writer_xruns_since_start

The number of xruns seen by the writer since the last start of processing.

• unsigned reader_xruns_since_start

The number of xruns seen by the reader since the last start of processing.

• unsigned writer_xruns_in_succession

The number of xruns seen by the writer in succession.

• unsigned reader_xruns_in_succession

The number of xruns seen by the reader in succession.

• unsigned maximum_writer_xruns_in_succession_before_stop

A limit to the number of xruns seen in succession during write before the data transmission

through the FIFO is stopped.

• unsigned maximum_reader_xruns_in_succession_before_stop

A limit to the number of xruns seen in succession during read before the data transmission

through the FIFO is stopped.

•mha_ﬁfo_t<T>::value_type null_data

The value used in place of missing data.

• unsigned startup_zeros

When processing starts, that is when both mha_drifter_ﬁfo_t<T>::reader_started (p. 443)

and mha_drifter_ﬁfo_t<T>::writer_started (p. 443) are true, then ﬁrst mha_drifter_ﬁfo←-

_t<T>::desired_ﬁll_count (p. 443) instances of mha_drifter_ﬁfo_t<T>::null_data (p. 444)

are delivered to the reader.

Additional Inherited Members

5.148.1 Detailed Description

template<class T>

class mha_drifter_ﬁfo_t<T>

A FIFO class for blocksize adaptation without Synchronization.

Features: delay concept (desired, minimum and maximum delay), drifting support by throwing

away data or inserting zeroes.

5.148 mha_drifter_ﬁfo_t<T>Class Template Reference 441

5.148.2 Constructor & Destructor Documentation

5.148.2.1 template<class T >mha_drifter_ﬁfo_t<T>::mha_drifter_ﬁfo_t (

unsigned min_ﬁll_count,

unsigned desired_ﬁll_count,

unsigned max_ﬁll_count )

Create drifter FIFO.

5.148.2.2 template<class T >mha_drifter_ﬁfo_t<T>::mha_drifter_ﬁfo_t (

unsigned min_ﬁll_count,

unsigned desired_ﬁll_count,

unsigned max_ﬁll_count,

const T & t)

Create drifter FIFO where all (initially unused) copies of T are initialized as copies of t.

5.148.3 Member Function Documentation

5.148.3.1 template<class T >void mha_drifter_ﬁfo_t<T>::write (

const T ∗data,

unsigned count )[virtual]

write data to ﬁfo

Sets writer_started (p. 443) to true.

When processing has started, i.e. both reader_started (p. 443) and writer_started (p. 443)

are true, write speciﬁed ammount of data to the ﬁfo. If there is not enough space available, then

the exceeding data is lost and the writer xrun counters are increased.

Processing is stopped when writer_xruns_in_succession (p. 444) exceeds maximum_←-

writer_xruns_in_succession_before_stop (p. 444).

Parameters

data Pointer to source data.

count Number of instances to copy

Reimplemented from mha_ﬁfo_t<T>(p. 454).

5.148.3.2 template<class T >void mha_drifter_ﬁfo_t<T>::read (

T∗buf,

unsigned count )[virtual]

Read data from ﬁfo.

442 CONTENTS

Sets reader_started (p. 443) to true.

When processing has started, i.e. both reader_started (p. 443) and writer_started (p. 443)

are true, then read speciﬁed ammount of data from the ﬁfo. As long as startup_zeros (p. 445)

is >0, null_data (p. 444) is delivered to the reader and startup_zeros (p. 445) is diminished.

Only when startup_zeros (p. 445) has reached 0, data is actually read from the ﬁfo's buffer.

If the read would cause the ﬁfo's ﬁll count to drop below minimum_ﬁll_count (p. 443), then

only so much data are read that minimum_ﬁll_count (p. 443) entries remain in the ﬁfo, the

missing data is replaced with null_data (p. 444), and the reader xrun counters are increased.

Processing is stopped when reader_xruns_in_succession (p. 444) exceeds maximum_←-

reader_xruns_in_succession_before_stop (p. 444).

Parameters

buf Pointer to the target buffer

count Number of instances to copy

Reimplemented from mha_ﬁfo_t<T>(p. 455).

5.148.3.3 template<class T >unsigned mha_drifter_ﬁfo_t<T>::get_ﬁll_count ( ) const

[virtual]

Return ﬁll_count, adding mha_drifter_ﬁfo_t<T>::startup_zeros (p. 445) to the number of

samples actually in the ﬁfo's buffer.

Reimplemented from mha_ﬁfo_t<T>(p. 455).

5.148.3.4 template<class T >unsigned mha_drifter_ﬁfo_t<T>::get_available_space ( ) const

[virtual]

Return available space, subtracting number of mha_drifter_ﬁfo_t<T>::startup_zeros

(p. 445) from the available_space actually present in the ﬁfo's buffer.

TODO: uncertain if this is a good idea.

Reimplemented from mha_ﬁfo_t<T>(p. 455).

5.148.3.5 template<class T >virtual unsigned mha_drifter_ﬁfo_t<T>::get_des_ﬁll_count ( )

const [inline],[virtual]

The desired ﬁll count of this ﬁfo.

5.148.3.6 template<class T >virtual unsigned mha_drifter_ﬁfo_t<T>::get_min_ﬁll_count ( )

const [inline],[virtual]

The minimum ﬁll count of this ﬁfo.

5.148 mha_drifter_ﬁfo_t<T>Class Template Reference 443

5.148.3.7 template<class T >void mha_drifter_ﬁfo_t<T>::stop ( ) [virtual]

Called by mha_drifter_ﬁfo_t<T>::read (p. 441) or mha_drifter_ﬁfo_t<T>::write (p. 441)

when their xrun in succession counter exceeds its limit.

Called by read (p. 441) or write (p. 441) when their xrun in succession counter exceeds its limit.

May also be called explicitly.

5.148.3.8 template<class T >void mha_drifter_ﬁfo_t<T>::starting ( ) [virtual]

Called by mha_drifter_ﬁfo_t<T>::read (p. 441) or mha_drifter_ﬁfo_t<T>::write (p. 441)

when the respective ﬂag (mha_drifter_ﬁfo_t<T>::reader_started (p. 443) or mha_drifter_←-

ﬁfo_t<T>::writer_started (p. 443)) is about to be toggled from false to true.

The ﬁfo's buffer is emptied, this method resets startup_zeros (p. 445) to desired_ﬁll_count

(p. 443), and it also resets reader_xruns_since_start (p. 444) and writer_xruns_since_start

(p. 444) to 0.

5.148.4 Member Data Documentation

5.148.4.1 template<class T >const unsigned mha_drifter_ﬁfo_t<T>::minimum_ﬁll_count

[private]

The minimum ﬁll count of this ﬁfo.

5.148.4.2 template<class T >const unsigned mha_drifter_ﬁfo_t<T>::desired_ﬁll_count

[private]

The desired ﬁll count of the ﬁfo.

The ﬁfo is initialized with this ammount of data when data transmission starts.

5.148.4.3 template<class T >bool mha_drifter_ﬁfo_t<T>::writer_started [private]

Flag set to true when write is called the ﬁrst time.

5.148.4.4 template<class T >bool mha_drifter_ﬁfo_t<T>::reader_started [private]

Flag set to true when read is called for the ﬁrst time.

5.148.4.5 template<class T >unsigned mha_drifter_ﬁfo_t<T>::writer_xruns_total [private]

The number of xruns seen by the writer since object instantiation.

444 CONTENTS

5.148.4.6 template<class T >unsigned mha_drifter_ﬁfo_t<T>::reader_xruns_total [private]

The number of xruns seen by the reader since object instantiation.

5.148.4.7 template<class T >unsigned mha_drifter_ﬁfo_t<T>::writer_xruns_since_start

[private]

The number of xruns seen by the writer since the last start of processing.

5.148.4.8 template<class T >unsigned mha_drifter_ﬁfo_t<T>::reader_xruns_since_start

[private]

The number of xruns seen by the reader since the last start of processing.

5.148.4.9 template<class T >unsigned mha_drifter_ﬁfo_t<T>::writer_xruns_in_succession

[private]

The number of xruns seen by the writer in succession.

Reset to 0 every time a write succeeds without xrun.

5.148.4.10 template<class T >unsigned mha_drifter_ﬁfo_t<T>::reader_xruns_in_succession

[private]

The number of xruns seen by the reader in succession.

Reset to 0 every time a read succeeds without xrun.

5.148.4.11 template<class T >unsigned mha_drifter_ﬁfo_t<T>::maximum_writer_xruns_in_←-

succession_before_stop [private]

A limit to the number of xruns seen in succession during write before the data transmission

through the FIFO is stopped.

5.148.4.12 template<class T >unsigned mha_drifter_ﬁfo_t<T>::maximum_reader_xruns_in_←-

succession_before_stop [private]

A limit to the number of xruns seen in succession during read before the data transmission

through the FIFO is stopped.

5.148.4.13 template<class T >mha_ﬁfo_t<T>::value_type mha_drifter_ﬁfo_t<T>::null_data

[private]

The value used in place of missing data.

5.149 MHA_Error Class Reference 445

5.148.4.14 template<class T >unsigned mha_drifter_ﬁfo_t<T>::startup_zeros [private]

When processing starts, that is when both mha_drifter_ﬁfo_t<T>::reader_started (p. 443)

and mha_drifter_ﬁfo_t<T>::writer_started (p. 443) are true, then ﬁrst mha_drifter_ﬁfo_t<←-

T>::desired_ﬁll_count (p. 443) instances of mha_drifter_ﬁfo_t<T>::null_data (p. 444) are

delivered to the reader.

These null_data (p. 444) instances are not transmitted through the ﬁfo because ﬁlling the ﬁfo

with enough null_data (p. 444) might not be realtime safe and this ﬁlling has to be initiated by

starting (p. 443) or stop (p. 443) (this implementation: starting (p. 443)) which are be called

with realtime constraints.

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.149 MHA_Error Class Reference

Error reporting exception class.

Inherits exception.

Public Member Functions

•MHA_Error (const char ∗ﬁle, int line, const char ∗fmt,...)

Create an instance of a MHA_Error (p. 445).

•MHA_Error (const MHA_Error &)

•MHA_Error &operator= (const MHA_Error &)

•∼MHA_Error () throw ()

• const char ∗get_msg () const

Return the error message without source position.

• const char ∗get_longmsg () const

Return the error message with source position.

• const char ∗what () const throw ()

overwrite std::execption::what()

Private Attributes

• char ∗msg

• char ∗longmsg

446 CONTENTS

5.149.1 Detailed Description

Error reporting exception class.

This class is used for error handling in the openMHA. It is used by the openMHA kernel and by

the openMHA toolbox library. Please note that exceptions should not be used accross ANSI-C

interfaces. It is necessary to catch exceptions within the library.

The MHA_Error (p. 445) class holds source ﬁle name, line number and an error message.

5.149.2 Constructor & Destructor Documentation

5.149.2.1 MHA_Error::MHA_Error (

const char ∗s_ﬁle,

int l,

const char ∗fmt,

... )

Create an instance of a MHA_Error (p. 445).

Parameters

s_ﬁle source ﬁle name (FILE)

lsource line (LINE)

fmt format string for error message (as in printf)

5.149.2.2 MHA_Error::MHA_Error (

const MHA_Error &p)

5.149.2.3 MHA_Error::∼MHA_Error ( ) throw )

5.149.3 Member Function Documentation

5.149.3.1 MHA_Error & MHA_Error::operator= (

const MHA_Error &p)

5.149.3.2 const char∗MHA_Error::get_msg ( ) const [inline]

Return the error message without source position.

5.149.3.3 const char∗MHA_Error::get_longmsg ( ) const [inline]

Return the error message with source position.

5.150 mha_ﬁfo_lw_t<T>Class Template Reference 447

5.149.3.4 const char∗MHA_Error::what ( ) const throw ) [inline]

overwrite std::execption::what()

5.149.4 Member Data Documentation

5.149.4.1 char∗MHA_Error::msg [private]

5.149.4.2 char∗MHA_Error::longmsg [private]

The documentation for this class was generated from the following ﬁles:

•mha_error.hh

•mha_error.cpp

5.150 mha_ﬁfo_lw_t<T>Class Template Reference

This FIFO uses locks to synchronize access.

Inheritance diagram for mha_ﬁfo_lw_t<T>:

mha_fifo_lw_t< T >

mha_fifo_t< T >

Public Member Functions

• virtual void write (const T ∗data, unsigned count)

write speciﬁed ammount of data to the ﬁfo.

• virtual void read (T ∗buf, unsigned count)

read data from ﬁfo.

•mha_ﬁfo_lw_t (unsigned max_ﬁll_count)

Create FIFO with ﬁxed buffer size.

• virtual ∼mha_ﬁfo_lw_t ()

release synchronization object

• virtual void set_error (unsigned index, MHA_Error ∗error)

Process waiting for more data or space should bail out, throwing this error.

448 CONTENTS

Private Attributes

•mha_ﬁfo_thread_platform_t ∗sync

platform speciﬁc thread synchronization

•MHA_Error ∗error [2]

If waiting for synchronization should be aborted then exception to be thrown by reader process

(index 0) or writer process (index 1) has to be placed here.

Additional Inherited Members

5.150.1 Detailed Description

template<class T>

class mha_ﬁfo_lw_t<T>

This FIFO uses locks to synchronize access.

Reading and writing can block until the operation can be executed.

5.150.2 Constructor & Destructor Documentation

5.150.2.1 template<class T >mha_ﬁfo_lw_t<T>::mha_ﬁfo_lw_t (

unsigned max_ﬁll_count )[explicit]

Create FIFO with ﬁxed buffer size.

5.150.2.2 template<class T >mha_ﬁfo_lw_t<T>::∼mha_ﬁfo_lw_t ( ) [virtual]

release synchronization object

5.150.3 Member Function Documentation

5.150.3.1 template<class T >void mha_ﬁfo_lw_t<T>::write (

const T ∗data,

unsigned count )[virtual]

write speciﬁed ammount of data to the ﬁfo.

If there is not enough space, then wait for more space.

5.150 mha_ﬁfo_lw_t<T>Class Template Reference 449

Parameters

data Pointer to source data.

count Number of instances to copy.

Exceptions

MHA_Error (p. 445)when detecting a deadlock situation.

Reimplemented from mha_ﬁfo_t<T>(p. 454).

5.150.3.2 template<class T >void mha_ﬁfo_lw_t<T>::read (

T∗buf,

unsigned count )[virtual]

read data from ﬁfo.

If there is not enough data, then wait for more data.

Parameters

buf Pointer to the target buffer.

count Number of instances to copy.

Exceptions

MHA_Error (p. 445)when detecting a deadlock situation.

Reimplemented from mha_ﬁfo_t<T>(p. 455).

5.150.3.3 template<class T >void mha_ﬁfo_lw_t<T>::set_error (

unsigned index,

MHA_Error ∗error )[virtual]

Process waiting for more data or space should bail out, throwing this error.

Parameters

index Use 0 for terminating reader, 1 for terminating writer.

error MHA_Error (p. 445) to be thrown

5.150.4 Member Data Documentation

450 CONTENTS

5.150.4.1 template<class T >mha_ﬁfo_thread_platform_t∗mha_ﬁfo_lw_t<T>::sync

[private]

platform speciﬁc thread synchronization

5.150.4.2 template<class T >MHA_Error∗mha_ﬁfo_lw_t<T>::error[2] [private]

If waiting for synchronization should be aborted then exception to be thrown by reader process

(index 0) or writer process (index 1) has to be placed here.

The documentation for this class was generated from the following ﬁles:

•mha_ﬁfo.h

•mha_ﬁfo.cpp

5.151 mha_ﬁfo_posix_threads_t Class Reference

Inheritance diagram for mha_ﬁfo_posix_threads_t:

mha_fifo_posix_threads_t

mha_fifo_thread_platform_t

Public Member Functions

•mha_ﬁfo_posix_threads_t ()

• virtual void aquire_mutex ()

• virtual void release_mutex ()

• virtual void wait_for_decrease ()

• virtual void wait_for_increase ()

• virtual void increment ()

• virtual void decrement ()

• virtual ∼mha_ﬁfo_posix_threads_t ()

5.151 mha_ﬁfo_posix_threads_t Class Reference 451

Private Attributes

• pthread_mutex_t mutex

• pthread_cond_t decrease_condition

• pthread_cond_t increase_condition

5.151.1 Constructor & Destructor Documentation

5.151.1.1 mha_ﬁfo_posix_threads_t::mha_ﬁfo_posix_threads_t ( ) [inline]

5.151.1.2 virtual mha_ﬁfo_posix_threads_t::∼mha_ﬁfo_posix_threads_t ( ) [inline],

[virtual]

5.151.2 Member Function Documentation

5.151.2.1 virtual void mha_ﬁfo_posix_threads_t::aquire_mutex ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 459).

5.151.2.2 virtual void mha_ﬁfo_posix_threads_t::release_mutex ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 459).

5.151.2.3 virtual void mha_ﬁfo_posix_threads_t::wait_for_decrease ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 459).

5.151.2.4 virtual void mha_ﬁfo_posix_threads_t::wait_for_increase ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 459).

5.151.2.5 virtual void mha_ﬁfo_posix_threads_t::increment ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 459).

5.151.2.6 virtual void mha_ﬁfo_posix_threads_t::decrement ( ) [inline],[virtual]

Implements mha_ﬁfo_thread_platform_t (p. 460).

452 CONTENTS

5.151.3 Member Data Documentation

5.151.3.1 pthread_mutex_t mha_ﬁfo_posix_threads_t::mutex [private]

5.151.3.2 pthread_cond_t mha_ﬁfo_posix_threads_t::decrease_condition [private]

5.151.3.3 pthread_cond_t mha_ﬁfo_posix_threads_t::increase_condition [private]

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.152 mha_ﬁfo_t<T>Class Template Reference

A FIFO class for blocksize adaptation Synchronization: None.

Inheritance diagram for mha_ﬁfo_t<T>:

mha_fifo_t< T >

mha_drifter_fifo_t< T > mha_fifo_lw_t< T >

Public Types

• typedef T value_type

The data type exchanged by this ﬁfo.

5.152 mha_ﬁfo_t<T>Class Template Reference 453

Public Member Functions

• virtual void write (const T ∗data, unsigned count)

write speciﬁed ammount of data to the ﬁfo.

• virtual void read (T ∗buf, unsigned count)

read data from ﬁfo

• virtual unsigned get_ﬁll_count () const

Read-only access to ﬁll_count.

• virtual unsigned get_available_space () const

Read-only access to available_space.

• virtual unsigned get_max_ﬁll_count () const

The capacity of this ﬁfo.

•mha_ﬁfo_t (unsigned max_ﬁll_count)

Create FIFO with ﬁxed buffer size.

•mha_ﬁfo_t (unsigned max_ﬁll_count, const T &t)

Create FIFO with ﬁxed buffer size, where all (initially unused) copies of T are initialized as

copies of t.

•mha_ﬁfo_t (const mha_ﬁfo_t &src)

Copy constructor.

• virtual ∼mha_ﬁfo_t ()

Destroy FIFO.

•mha_ﬁfo_t<T>&operator= (const mha_ﬁfo_t<T>&)

Assignment operator.

Protected Member Functions

• void clear ()

Empty the ﬁfo at once.

Private Attributes

• const unsigned max_ﬁll_count

The maximum ﬁll count of this FIFO.

• T ∗buf

The memory allocated to store the data.

• T ∗write_ptr

points to location where to write next

• const T ∗read_ptr

points to location where to read next

• bool buf_uses_placement_new

wether buf was allocated using placement new or array new.

454 CONTENTS

5.152.1 Detailed Description

template<class T>

class mha_ﬁfo_t<T>

A FIFO class for blocksize adaptation Synchronization: None.

Use external synchronisation or synchronization in inheriting class.

5.152.2 Member Typedef Documentation

5.152.2.1 template<class T>typedef T mha_ﬁfo_t<T>::value_type

The data type exchanged by this ﬁfo.

5.152.3 Constructor & Destructor Documentation

5.152.3.1 template<class T >mha_ﬁfo_t<T>::mha_ﬁfo_t (

unsigned max_ﬁll_count )[explicit]

Create FIFO with ﬁxed buffer size.

5.152.3.2 template<class T>mha_ﬁfo_t<T>::mha_ﬁfo_t (

unsigned max_ﬁll_count,

const T & t)

Create FIFO with ﬁxed buffer size, where all (initially unused) copies of T are initialized as

copies of t.

5.152.3.3 template<class T>mha_ﬁfo_t<T>::mha_ﬁfo_t (

const mha_ﬁfo_t<T>&src )

Copy constructor.

5.152.3.4 template<class T >mha_ﬁfo_t<T>::∼mha_ﬁfo_t ( ) [virtual]

Destroy FIFO.

5.152.4 Member Function Documentation

5.152.4.1 template<class T>void mha_ﬁfo_t<T>::write (

const T ∗data,

unsigned count )[virtual]

write speciﬁed ammount of data to the ﬁfo.

5.152 mha_ﬁfo_t<T>Class Template Reference 455

Parameters

data Pointer to source data.

count Number of instances to copy

Exceptions

MHA_Error (p. 445)when there is not enough space available.

Reimplemented in mha_ﬁfo_lw_t<T>(p. 448), and mha_drifter_ﬁfo_t<T>(p. 441).

5.152.4.2 template<class T>void mha_ﬁfo_t<T>::read (

T∗buf,

unsigned count )[virtual]

read data from ﬁfo

Parameters

buf Pointer to the target buffer

count Number of instances to copy

Exceptions

MHA_Error (p. 445)when there is not enough data available.

Reimplemented in mha_ﬁfo_lw_t<T>(p. 449), and mha_drifter_ﬁfo_t<T>(p. 441).

5.152.4.3 template<class T >unsigned mha_ﬁfo_t<T>::get_ﬁll_count ( ) const [virtual]

Read-only access to ﬁll_count.

Reimplemented in mha_drifter_ﬁfo_t<T>(p. 442).

5.152.4.4 template<class T >unsigned mha_ﬁfo_t<T>::get_available_space ( ) const

[virtual]

Read-only access to available_space.

Reimplemented in mha_drifter_ﬁfo_t<T>(p. 442).

5.152.4.5 template<class T>virtual unsigned mha_ﬁfo_t<T>::get_max_ﬁll_count ( ) const

[inline],[virtual]

The capacity of this ﬁfo.

456 CONTENTS

5.152.4.6 template<class T>mha_ﬁfo_t<T>&mha_ﬁfo_t<T>::operator= (

const mha_ﬁfo_t<T>&src )

Assignment operator.

5.152.4.7 template<class T>void mha_ﬁfo_t<T>::clear ( ) [inline],[protected]

Empty the ﬁfo at once.

Should be called by the reader, or when the reader is inactive.

5.152.5 Member Data Documentation

5.152.5.1 template<class T>const unsigned mha_ﬁfo_t<T>::max_ﬁll_count [private]

The maximum ﬁll count of this FIFO.

5.152.5.2 template<class T>T∗mha_ﬁfo_t<T>::buf [private]

The memory allocated to store the data.

max_ﬁll_count + 1 locations are allocated: At least one location is always unused, because we

have max_ﬁll_count + 1 possible ﬁllcounts [0:max_ﬁll_count] that we need to distinguish.

5.152.5.3 template<class T>T∗mha_ﬁfo_t<T>::write_ptr [private]

points to location where to write next

5.152.5.4 template<class T>const T∗mha_ﬁfo_t<T>::read_ptr [private]

points to location where to read next

5.152.5.5 template<class T>bool mha_ﬁfo_t<T>::buf_uses_placement_new [private]

wether buf was allocated using placement new or array new.

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.153 mha_ﬁfo_thread_guard_t Class Reference

Simple Mutex Guard Class.

5.154 mha_ﬁfo_thread_platform_t Class Reference 457

Public Member Functions

•mha_ﬁfo_thread_guard_t (mha_ﬁfo_thread_platform_t ∗sync)

•∼mha_ﬁfo_thread_guard_t ()

Private Attributes

•mha_ﬁfo_thread_platform_t ∗sync

5.153.1 Detailed Description

Simple Mutex Guard Class.

5.153.2 Constructor & Destructor Documentation

5.153.2.1 mha_ﬁfo_thread_guard_t::mha_ﬁfo_thread_guard_t (

mha_ﬁfo_thread_platform_t ∗sync )[inline]

5.153.2.2 mha_ﬁfo_thread_guard_t::∼mha_ﬁfo_thread_guard_t ( ) [inline]

5.153.3 Member Data Documentation

5.153.3.1 mha_ﬁfo_thread_platform_t∗mha_ﬁfo_thread_guard_t::sync [private]

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.154 mha_ﬁfo_thread_platform_t Class Reference

Abstract base class for synchronizing multithreaded (producer/consumer) ﬁfo operations.

Inheritance diagram for mha_ﬁfo_thread_platform_t:

mha_fifo_thread_platform_t

mha_fifo_posix_threads_t

458 CONTENTS

Public Member Functions

• virtual void aquire_mutex ()=0

Calling thread waits until it aquires the lock.

• virtual void release_mutex ()=0

Calling thread releases the lock.

• virtual void wait_for_decrease ()=0

Calling producer thread must own the lock.

• virtual void wait_for_increase ()=0

Calling consumer thread must own the lock.

• virtual void increment ()=0

To be called by producer thread after producing.

• virtual void decrement ()=0

To be called by consumer thread after consuming.

• virtual ∼mha_ﬁfo_thread_platform_t ()

Make destructor virtual.

•mha_ﬁfo_thread_platform_t ()

Make default constructor accessible.

Private Member Functions

•mha_ﬁfo_thread_platform_t (const mha_ﬁfo_thread_platform_t &)

•mha_ﬁfo_thread_platform_t &operator= (const mha_ﬁfo_thread_platform_t &)

5.154.1 Detailed Description

Abstract base class for synchronizing multithreaded (producer/consumer) ﬁfo operations.

Works only with single producer and single consumer.

5.154.2 Constructor & Destructor Documentation

5.154.2.1 virtual mha_ﬁfo_thread_platform_t::∼mha_ﬁfo_thread_platform_t ( ) [inline],

[virtual]

Make destructor virtual.

5.154.2.2 mha_ﬁfo_thread_platform_t::mha_ﬁfo_thread_platform_t (

const mha_ﬁfo_thread_platform_t & ) [private]

5.154.2.3 mha_ﬁfo_thread_platform_t::mha_ﬁfo_thread_platform_t ( ) [inline]

Make default constructor accessible.

5.154 mha_ﬁfo_thread_platform_t Class Reference 459

5.154.3 Member Function Documentation

5.154.3.1 virtual void mha_ﬁfo_thread_platform_t::aquire_mutex ( ) [pure virtual]

Calling thread waits until it aquires the lock.

Must not be called when the lock is already aquired.

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

5.154.3.2 virtual void mha_ﬁfo_thread_platform_t::release_mutex ( ) [pure virtual]

Calling thread releases the lock.

May only be called when lock is owned.

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

5.154.3.3 virtual void mha_ﬁfo_thread_platform_t::wait_for_decrease ( ) [pure virtual]

Calling producer thread must own the lock.

Method releases lock, and waits for consumer thread to call decrease(). Then reaquires lock

and returns

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

5.154.3.4 virtual void mha_ﬁfo_thread_platform_t::wait_for_increase ( ) [pure virtual]

Calling consumer thread must own the lock.

Method releases lock, and waits for producer thread to call increase(). Then reaquires lock and

returns

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

5.154.3.5 virtual void mha_ﬁfo_thread_platform_t::increment ( ) [pure virtual]

To be called by producer thread after producing.

Producer thread needs to own the lock to call this method.

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

460 CONTENTS

5.154.3.6 virtual void mha_ﬁfo_thread_platform_t::decrement ( ) [pure virtual]

To be called by consumer thread after consuming.

Consumer thread needs to own the lock to call this method.

Implemented in mha_ﬁfo_posix_threads_t (p. 451).

5.154.3.7 mha_ﬁfo_thread_platform_t& mha_ﬁfo_thread_platform_t::operator= (

const mha_ﬁfo_thread_platform_t & ) [private]

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.155 mha_real_test_array_t Struct Reference

Public Attributes

•mha_real_t r [4]

5.155.1 Member Data Documentation

5.155.1.1 mha_real_t mha_real_test_array_t::r[4]

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.156 mha_rt_ﬁfo_element_t<T>Class Template Reference

Object wrapper for mha_rt_ﬁfo_t (p. 462).

Public Member Functions

•mha_rt_ﬁfo_element_t (T ∗data)

Constructor.

•∼mha_rt_ﬁfo_element_t ()

5.156 mha_rt_ﬁfo_element_t<T>Class Template Reference 461

Public Attributes

•mha_rt_ﬁfo_element_t<T>∗next

Pointer to next ﬁfo element. NULL for the last (newest) ﬁfo element.

• bool abandonned

Indicates that this element will no longer be used and may be deleted.

• T ∗data

Pointer to user data.

5.156.1 Detailed Description

template<class T>

class mha_rt_ﬁfo_element_t<T>

Object wrapper for mha_rt_ﬁfo_t (p. 462).

5.156.2 Constructor & Destructor Documentation

5.156.2.1 template<class T >mha_rt_ﬁfo_element_t<T>::mha_rt_ﬁfo_element_t (

T∗data )[inline]

Constructor.

This element assumes ownership of user data.

Parameters

data User data. Has to be allocated on the heap with standard operator new, because it

will be deleted in this element's destructor.

5.156.2.2 template<class T >mha_rt_ﬁfo_element_t<T>::∼mha_rt_ﬁfo_element_t ( )

[inline]

5.156.3 Member Data Documentation

5.156.3.1 template<class T >mha_rt_ﬁfo_element_t<T>∗mha_rt_ﬁfo_element_t<T>::next

Pointer to next ﬁfo element. NULL for the last (newest) ﬁfo element.

5.156.3.2 template<class T >bool mha_rt_ﬁfo_element_t<T>::abandonned

Indicates that this element will no longer be used and may be deleted.

462 CONTENTS

5.156.3.3 template<class T >T∗mha_rt_ﬁfo_element_t<T>::data

Pointer to user data.

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.157 mha_rt_ﬁfo_t<T>Class Template Reference

Template class for thread safe, half real time safe ﬁfo without explixit locks.

Public Member Functions

•mha_rt_ﬁfo_t ()

Construct empty ﬁfo.

•∼mha_rt_ﬁfo_t ()

Destructor will delete all data currently in the ﬁfo.

• T ∗poll ()

Retrieve the latest element in the Fifo.

• T ∗poll_1 ()

Retrieve the next element in the Fifo, if there is one, and mark the previous element as aban-

donned.

• void push (T ∗data)

Add element to the Fifo.

Private Member Functions

• void remove_abandonned ()

Deletes abandonned elements.

• void remove_all ()

Deletes all elements.

Private Attributes

•mha_rt_ﬁfo_element_t<T>∗root

The ﬁrst element in the ﬁfo. Deleting elements starts here.

•mha_rt_ﬁfo_element_t<T>∗current

The element most recently returned by poll (p. 463) or poll_1 (p. 464).

5.157 mha_rt_ﬁfo_t<T>Class Template Reference 463

5.157.1 Detailed Description

template<class T>

class mha_rt_ﬁfo_t<T>

Template class for thread safe, half real time safe ﬁfo without explixit locks.

Reading from this ﬁfo is realtime safe, writing to it is not. This ﬁfo is designed for objects that

were constructed on the heap. It assumes ownership of these objects and calls delete on them

when they are no longer used. Objects remain inside the Fifo while being used by the reader.

A new ﬁfo element is inserted by using push (p. 464). The push operation is not real time safe,

it allocates and deallocates memory. The latest element is retrieved by calling poll (p. 463).

This operation will skip ﬁfo elements if more than one push (p. 464) has been occured since

the last poll. To avoid skipping, call the poll_1 (p. 464) operation instead.

5.157.2 Constructor & Destructor Documentation

5.157.2.1 template<class T >mha_rt_ﬁfo_t<T>::mha_rt_ﬁfo_t ( ) [inline]

Construct empty ﬁfo.

5.157.2.2 template<class T >mha_rt_ﬁfo_t<T>::∼mha_rt_ﬁfo_t ( ) [inline]

Destructor will delete all data currently in the ﬁfo.

5.157.3 Member Function Documentation

5.157.3.1 template<class T >T∗mha_rt_ﬁfo_t<T>::poll ( ) [inline]

Retrieve the latest element in the Fifo.

Will skip ﬁfo elements if more than one element has been added since last poll invocation. Will

return the same element as on last call if no elements have been added in the mean time.

Marks former elements as abandonned.

Returns

The latest element in this Fifo. Returns NULL if the Fifo is empty.

464 CONTENTS

5.157.3.2 template<class T >T∗mha_rt_ﬁfo_t<T>::poll_1 ( ) [inline]

Retrieve the next element in the Fifo, if there is one, and mark the previous element as aban-

donned.

Else, if there is no newer element, returns the same element as on last poll() (p. 463) or poll_1()

(p. 464) invocation.

Returns

The next element in this Fifo, if there is one, or the same as before. Returns NULL if the

Fifo is empty.

5.157.3.3 template<class T >void mha_rt_ﬁfo_t<T>::push (

T∗data )[inline]

Add element to the Fifo.

Deletes abandonned elements in the ﬁfo.

Parameters

data The new user data to place at the end of the ﬁfo. After this invocation, the ﬁfo is the

owner of this object and will delete it when it is no longer used. data must have been

allocated on the heap with standard operator new.

5.157.3.4 template<class T >void mha_rt_ﬁfo_t<T>::remove_abandonned ( ) [inline],

[private]

Deletes abandonned elements.

5.157.3.5 template<class T >void mha_rt_ﬁfo_t<T>::remove_all ( ) [inline],[private]

Deletes all elements.

5.157.4 Member Data Documentation

5.157.4.1 template<class T >mha_rt_ﬁfo_element_t<T>∗mha_rt_ﬁfo_t<T>::root

[private]

The ﬁrst element in the ﬁfo. Deleting elements starts here.

5.158 mha_spec_t Struct Reference 465

5.157.4.2 template<class T >mha_rt_ﬁfo_element_t<T>∗mha_rt_ﬁfo_t<T>::current

[private]

The element most recently returned by poll (p. 463) or poll_1 (p. 464).

Searching for new elements starts here.

The documentation for this class was generated from the following ﬁle:

•mha_ﬁfo.h

5.158 mha_spec_t Struct Reference

Spectrum signal structure.

Inheritance diagram for mha_spec_t:

mha_spec_t

MHASignal::spectrum_t

fshift_hilbert::hilbert

_shifter_t MHA_AC::spectrum_t

MHAMultiSrc::spectrum_t wave2spec_t

Public Attributes

•mha_complex_t ∗buf

signal buffer

• unsigned int num_channels

number of channels

• unsigned int num_frames

number of frames in each channel

•mha_channel_info_t ∗channel_info

detailed channel description

466 CONTENTS

5.158.1 Detailed Description

Spectrum signal structure.

This structure contains the short time fourier transform output of the windowed input signal.

The member num_frames describes the number of frequency bins in each channel. For an

even FFT length N, this is N/2+1. With odd FFT lengths, it is (N+ 1)/2. The imaginary part

of the ﬁrst bin is zero. For even FFT lengths, also the imaginary part at the Nyquist frequency

is zero.

buf[k].re Re(0) Re(1) Re(2) Re(3) Re(4) ... Re(n/2−1) Re(n/2)

buf[k].im Im(1) Im(2) Im(3) Im(4) ... Im(n/2−1)

k0 1 2 3 4 n/2−1 n/2

Figure 4 Data order of FFT spectrum.

5.158.2 Member Data Documentation

5.158.2.1 mha_complex_t∗mha_spec_t::buf

signal buffer

5.158.2.2 unsigned int mha_spec_t::num_channels

number of channels

5.158.2.3 unsigned int mha_spec_t::num_frames

number of frames in each channel

5.158.2.4 mha_channel_info_t∗mha_spec_t::channel_info

detailed channel description

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.159 mha_stash_environment_variable_t Class Reference

This class changes the value of an environment variable when constructed and restores the

original state of the environment variable when destroyed.

5.159 mha_stash_environment_variable_t Class Reference 467

Public Member Functions

•mha_stash_environment_variable_t (const std::string &variable_name, const std←-

::string &new_content)

•∼mha_stash_environment_variable_t ()

Private Attributes

• const bool existed_before

Flag indicates if the environment variable existed before constructor.

• const std::string variable_name

Name of environment variable.

• const std::string original_content

Content of environment variable before constructor executed.

5.159.1 Detailed Description

This class changes the value of an environment variable when constructed and restores the

original state of the environment variable when destroyed.

Can be used for testing functionality related to environment variables.

Todo Move to collection of unit-test support classes when we have one.

5.159.2 Constructor & Destructor Documentation

5.159.2.1 mha_stash_environment_variable_t::mha_stash_environment_variable_t (

const std::string & variable_name,

const std::string & new_content )[inline]

5.159.2.2 mha_stash_environment_variable_t::∼mha_stash_environment_variable_t ( ) [inline]

5.159.3 Member Data Documentation

5.159.3.1 const bool mha_stash_environment_variable_t::existed_before [private]

Flag indicates if the environment variable existed before constructor.

5.159.3.2 const std::string mha_stash_environment_variable_t::variable_name [private]

Name of environment variable.

468 CONTENTS

5.159.3.3 const std::string mha_stash_environment_variable_t::original_content [private]

Content of environment variable before constructor executed.

The documentation for this class was generated from the following ﬁle:

•mha_os.h

5.160 MHA_TCP::Async_Notify Class Reference

Portable Multiplexable cross-thread notiﬁcation.

Inheritance diagram for MHA_TCP::Async_Notify:

MHA_TCP::Async_Notify

MHA_TCP::Wakeup_Event

Public Member Functions

•Async_Notify ()

• virtual void reset ()

• virtual void set ()

• virtual ∼Async_Notify ()

Private Attributes

• int pipe [2]

Additional Inherited Members

5.160.1 Detailed Description

Portable Multiplexable cross-thread notiﬁcation.

5.161 MHA_TCP::Client Class Reference 469

5.160.2 Constructor & Destructor Documentation

5.160.2.1 Async_Notify::Async_Notify ( )

5.160.2.2 Async_Notify::∼Async_Notify ( ) [virtual]

5.160.3 Member Function Documentation

5.160.3.1 void Async_Notify::reset ( ) [virtual]

Reimplemented from MHA_TCP::Wakeup_Event (p. 495).

5.160.3.2 void Async_Notify::set ( ) [virtual]

5.160.4 Member Data Documentation

5.160.4.1 int MHA_TCP::Async_Notify::pipe[2] [private]

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.161 MHA_TCP::Client Class Reference

A portable class for a tcp client connections.

Inheritance diagram for MHA_TCP::Client:

MHA_TCP::Client

MHA_TCP::Connection

470 CONTENTS

Public Member Functions

•Client (const std::string &host, unsigned short port)

Constructor connects to host, port via TCP.

•Client (const std::string &host, unsigned short port, Timeout_Watcher &timeout_←-

watcher)

Constructor connects to host, port via TCP, using a timeout.

Additional Inherited Members

5.161.1 Detailed Description

A portable class for a tcp client connections.

5.161.2 Constructor & Destructor Documentation

5.161.2.1 Client::Client (

const std::string & host,

unsigned short port )

Constructor connects to host, port via TCP.

Parameters

host The hostname of the TCP Server (p. 481).

port The port or the TCP Server (p. 481).

5.161.2.2 Client::Client (

const std::string & host,

unsigned short port,

Timeout_Watcher &timeout_watcher )

Constructor connects to host, port via TCP, using a timeout.

Parameters

host The hostname of the TCP Server (p. 481).

port The port or the TCP Server (p. 481).

timeout_watcher an Event watcher that implements a timeout.

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

5.162 MHA_TCP::Connection Class Reference 471

•mha_tcp.cpp

5.162 MHA_TCP::Connection Class Reference

Connection (p. 471) handles Communication between client and server, is used on both sides.

Inheritance diagram for MHA_TCP::Connection:

MHA_TCP::Connection

MHA_TCP::Client

Public Member Functions

•Sockread_Event ∗get_read_event ()

•Sockwrite_Event ∗get_write_event ()

• std::string get_peer_address ()

Get peer's IP Address.

• unsigned short get_peer_port ()

Get peer's TCP port.

•SOCKET get_fd () const

Return the (protected) ﬁle descriptor of the connection.

• virtual ∼Connection ()

Destructor closes the underlying ﬁle descriptor.

• bool eof ()

Checks if the peer has closed the connection.

• bool can_read_line (char delim= '\n')

Checks if a full line of text has arrived by now.

• bool can_read_bytes (unsigned howmany)

Checks if the speciﬁed ammount of data can be read.

• std::string read_line (char delim= '\n')

Reads a single line of data from the socket.

• std::string read_bytes (unsigned howmany)

Reads the speciﬁed ammount of dat from the socket.

• void try_write (const std::string &data="")

472 CONTENTS

Adds data to the internal "outgoing" buffer, and then tries to write as much data from that buffer

to the socket as possible without blocking.

• void write (const std::string &data="")

Adds data to the internal "outgoing" buffer, and then writes that that buffer to the socket, re-

gardless of blocking.

• bool needs_write ()

Checks if the internal "outgoing" buffer contains data.

• unsigned buffered_incoming_bytes () const

Returns the number of bytes in the internal "incoming" buffer.

• unsigned buffered_outgoing_bytes () const

Returns the number of bytes in the internal "outgoing" buffer.

Protected Member Functions

•Connection (SOCKET _fd)

Create a connection instance from a socket ﬁledescriptor.

Protected Attributes

•SOCKET fd

The ﬁle descriptor of the TCP Socket.

Private Member Functions

• void init_peer_data ()

determine peer address and port

• bool can_sysread ()

Determine wether at least 1 byte can be read without blocking.

• bool can_syswrite ()

Determine wether at least 1 byte can be written without blocking.

• std::string sysread (unsigned bytes)

Call the system's read function and try to read bytes.

• std::string syswrite (const std::string &data)

Call the system's write function and try to write all characters in the string data.

Private Attributes

• std::string outbuf

• std::string inbuf

•Sockread_Event ∗read_event

•Sockwrite_Event ∗write_event

• bool closed

• struct sockaddr_in peer_addr

5.162 MHA_TCP::Connection Class Reference 473

5.162.1 Detailed Description

Connection (p. 471) handles Communication between client and server, is used on both sides.

5.162.2 Constructor & Destructor Documentation

5.162.2.1 MHA_TCP::Connection::Connection (

SOCKET _fd )[protected]

Create a connection instance from a socket ﬁledescriptor.

Parameters

←-

_←-

The ﬁle descriptor of the TCP Socket. This ﬁle descriptor is closed again in the

destructor.

Exceptions

MHA_Error (p. 445)If the ﬁle descriptor is <0.

5.162.2.2 Connection::∼Connection ( ) [virtual]

Destructor closes the underlying ﬁle descriptor.

5.162.3 Member Function Documentation

5.162.3.1 void MHA_TCP::Connection::init_peer_data ( ) [private]

determine peer address and port

5.162.3.2 bool Connection::can_sysread ( ) [private]

Determine wether at least 1 byte can be read without blocking.

5.162.3.3 bool Connection::can_syswrite ( ) [private]

Determine wether at least 1 byte can be written without blocking.

5.162.3.4 std::string Connection::sysread (

unsigned bytes )[private]

Call the system's read function and try to read bytes.

This will block in a situation where can_sysread returns false.

474 CONTENTS

Parameters

bytes The desired number of characters.

Returns

The characters read from the socket. The result may have fewer characters than speciﬁed

by bytes. If the result is an empty string, then the socket has been closed by the peer.

5.162.3.5 std::string Connection::syswrite (

const std::string & data )[private]

Call the system's write function and try to write all characters in the string data.

May write fewer characters, but will at least write one character.

Parameters

data A string of characters to write to the socket.

Returns

The rest of the characters that have not yet been written.

5.162.3.6 Sockread_Event ∗Connection::get_read_event ( )

5.162.3.7 Sockwrite_Event ∗Connection::get_write_event ( )

5.162.3.8 std::string Connection::get_peer_address ( )

Get peer's IP Address.

5.162.3.9 unsigned short Connection::get_peer_port ( )

Get peer's TCP port.

5.162.3.10 SOCKET MHA_TCP::Connection::get_fd ( ) const [inline]

Return the (protected) ﬁle descriptor of the connection.

Will be required for SSL.

5.162 MHA_TCP::Connection Class Reference 475

5.162.3.11 bool Connection::eof ( )

Checks if the peer has closed the connection.

As a side effect, this method ﬁlls the internal "incoming" buffer if it was empty and the socket is

readable and not eof.

5.162.3.12 bool Connection::can_read_line (

char delim = '\n' )

Checks if a full line of text has arrived by now.

This method reads data from the socket into the internal "incoming" buffer if it can be done

without blocking.

Parameters

delim The line delimiter.

Returns

true if at least one full line of text is present in the internal buffer after this method call,

false otherwise.

5.162.3.13 bool Connection::can_read_bytes (

unsigned howmany )

Checks if the speciﬁed ammount of data can be read.

This method reads data from the socket into an internal "incoming" buffer if it can be done

without blocking.

Parameters

howmany The number of bytes that the caller wants to have checked.

Returns

true if at least the speciﬁed ammount of data is present in the internal buffer after this

method call, false otherwise

5.162.3.14 std::string Connection::read_line (

char delim = '\n' )

Reads a single line of data from the socket.

Blocks if necessary.

476 CONTENTS

Parameters

delim The line delimiter.

Returns

The string of characters in this line, including the trailing delimiter. The delimiter may be

missing if the last line before EOF does not have a delimiter.

5.162.3.15 std::string Connection::read_bytes (

unsigned howmany )

Reads the speciﬁed ammount of dat from the socket.

Blocks if necessary.

Parameters

howmany The number of bytes to read.

Returns

The string of characters read. The string may be shorter if EOF is encountered.

5.162.3.16 void Connection::try_write (

const std::string & data = "" )

Adds data to the internal "outgoing" buffer, and then tries to write as much data from that buffer

to the socket as possible without blocking.

Parameters

data data to send over the socket.

5.162.3.17 void Connection::write (

const std::string & data = "" )

Adds data to the internal "outgoing" buffer, and then writes that that buffer to the socket, re-

gardless of blocking.

Parameters

data data to send over the socket.

5.162 MHA_TCP::Connection Class Reference 477

5.162.3.18 bool Connection::needs_write ( )

Checks if the internal "outgoing" buffer contains data.

5.162.3.19 unsigned Connection::buffered_incoming_bytes ( ) const

Returns the number of bytes in the internal "incoming" buffer.

5.162.3.20 unsigned Connection::buffered_outgoing_bytes ( ) const

Returns the number of bytes in the internal "outgoing" buffer.

5.162.4 Member Data Documentation

5.162.4.1 std::string MHA_TCP::Connection::outbuf [private]

5.162.4.2 std::string MHA_TCP::Connection::inbuf [private]

5.162.4.3 Sockread_Event∗MHA_TCP::Connection::read_event [private]

5.162.4.4 Sockwrite_Event∗MHA_TCP::Connection::write_event [private]

5.162.4.5 bool MHA_TCP::Connection::closed [private]

5.162.4.6 struct sockaddr_in MHA_TCP::Connection::peer_addr [private]

5.162.4.7 SOCKET MHA_TCP::Connection::fd [protected]

The ﬁle descriptor of the TCP Socket.

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

478 CONTENTS

5.163 MHA_TCP::Event_Watcher Class Reference

OS-independent event watcher, uses select on Unix and WaitForMultipleObjects on Windows.

Inheritance diagram for MHA_TCP::Event_Watcher:

MHA_TCP::Event_Watcher

MHA_TCP::Timeout_Watcher

Public Types

• typedef std::set<Wakeup_Event ∗>Events

• typedef std::set<Wakeup_Event ∗>::iterator iterator

Public Member Functions

• void observe (Wakeup_Event ∗event)

Add an event to this observer.

• void ignore (Wakeup_Event ∗event)

Remove an event from this observer.

•std::set<Wakeup_Event ∗>wait ()

\ Wait for some event to occur.

• virtual ∼Event_Watcher ()

Private Attributes

•std::set<Wakeup_Event ∗>events

The list of events to watch.

5.163.1 Detailed Description

OS-independent event watcher, uses select on Unix and WaitForMultipleObjects on Windows.

5.163 MHA_TCP::Event_Watcher Class Reference 479

5.163.2 Member Typedef Documentation

5.163.2.1 typedef std::set<Wakeup_Event∗>MHA_TCP::Event_Watcher::Events

5.163.2.2 typedef std::set<Wakeup_Event∗>::iterator MHA_TCP::Event_Watcher::iterator

5.163.3 Constructor & Destructor Documentation

5.163.3.1 Event_Watcher::∼Event_Watcher ( ) [virtual]

5.163.4 Member Function Documentation

5.163.4.1 void Event_Watcher::observe (

Wakeup_Event ∗event )

Add an event to this observer.

5.163.4.2 void Event_Watcher::ignore (

Wakeup_Event ∗event )

Remove an event from this observer.

5.163.4.3 std::set<Wakeup_Event ∗>Event_Watcher::wait ( )

\ Wait for some event to occur.

Return all events that are ready

5.163.5 Member Data Documentation

5.163.5.1 std::set<Wakeup_Event∗>MHA_TCP::Event_Watcher::events [private]

The list of events to watch.

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

480 CONTENTS

5.164 MHA_TCP::OS_EVENT_TYPE Struct Reference

Public Types

Public Attributes

• enum MHA_TCP::OS_EVENT_TYPE:: { ... } mode

• union {

int fd

double timeout

};

5.164.1 Member Enumeration Documentation

5.164.1.1 anonymous enum

Enumerator

5.164.2 Member Data Documentation

5.164.2.1 enum { ... } MHA_TCP::OS_EVENT_TYPE::mode

5.164.2.2 int MHA_TCP::OS_EVENT_TYPE::fd

5.164.2.3 double MHA_TCP::OS_EVENT_TYPE::timeout

5.164.2.4 union { ... }

The documentation for this struct was generated from the following ﬁle:

•mha_tcp.hh

5.165 MHA_TCP::Server Class Reference 481

5.165 MHA_TCP::Server Class Reference

Public Member Functions

•Server (unsigned short port=0, const std::string &iface="0.0.0.0")

Create a TCP server socket.

•Server (const std::string &iface, unsigned short port=0)

Create a TCP server socket.

•∼Server ()

Close the TCP server socket.

• std::string get_interface () const

Get the name given in the constructor for the network interface.

• unsigned short get_port () const

Get the port that the TCP server socket currently listens to.

•Sockaccept_Event ∗get_accept_event ()

Produces an event that can be observed by an Event_Watcher (p. 478).

•Connection ∗accept ()

Accept an incoming connection.

•Connection ∗try_accept ()

Accept an incoming connection if it can be done without blocking.

Private Member Functions

• void initialize (const std::string &iface, unsigned short port)

Private Attributes

• sockaddr_in sock_addr

•SOCKET serversocket

• std::string iface

• unsigned short port

•Sockaccept_Event ∗accept_event

5.165.1 Constructor & Destructor Documentation

5.165.1.1 Server::Server (

unsigned short port = 0,

const std::string & iface = "0.0.0.0" )

Create a TCP server socket.

482 CONTENTS

Parameters

port The TCP port to listen to.

iface The network interface to bind to.

5.165.1.2 Server::Server (

const std::string & iface,

unsigned short port = 0)

Create a TCP server socket.

Parameters

port The TCP port to listen to.

iface The network interface to bind to.

5.165.1.3 Server::∼Server ( )

Close the TCP server socket.

5.165.2 Member Function Documentation

5.165.2.1 void Server::initialize (

const std::string & iface,

unsigned short port )[private]

5.165.2.2 std::string Server::get_interface ( ) const

Get the name given in the constructor for the network interface.

5.165.2.3 unsigned short Server::get_port ( ) const

Get the port that the TCP server socket currently listens to.

5.165.2.4 Sockaccept_Event ∗Server::get_accept_event ( )

Produces an event that can be observed by an Event_Watcher (p. 478).

This event signals incoming connections that can be accepted.

5.166 MHA_TCP::sock_init_t Class Reference 483

5.165.2.5 Connection ∗Server::accept ( )

Accept an incoming connection.

blocks if necessary.

Returns

The new TCP connection. The connection has to be deleted by the caller.

5.165.2.6 Connection ∗Server::try_accept ( )

Accept an incoming connection if it can be done without blocking.

Returns

The new TCP connection or 0 if there is no immediate connection. The connection has to

be deleted by the caller.

5.165.3 Member Data Documentation

5.165.3.1 sockaddr_in MHA_TCP::Server::sock_addr [private]

5.165.3.2 SOCKET MHA_TCP::Server::serversocket [private]

5.165.3.3 std::string MHA_TCP::Server::iface [private]

5.165.3.4 unsigned short MHA_TCP::Server::port [private]

5.165.3.5 Sockaccept_Event∗MHA_TCP::Server::accept_event [private]

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.166 MHA_TCP::sock_init_t Class Reference

Public Member Functions

•sock_init_t ()

484 CONTENTS

5.166.1 Constructor & Destructor Documentation

5.166.1.1 MHA_TCP::sock_init_t::sock_init_t ( ) [inline]

The documentation for this class was generated from the following ﬁle:

•mha_tcp.cpp

5.167 MHA_TCP::Sockaccept_Event Class Reference

Inheritance diagram for MHA_TCP::Sockaccept_Event:

MHA_TCP::Sockaccept

_Event

MHA_TCP::Wakeup_Event

Public Member Functions

•Sockaccept_Event (SOCKET)

Additional Inherited Members

5.167.1 Constructor & Destructor Documentation

5.167.1.1 MHA_TCP::Sockaccept_Event::Sockaccept_Event (

SOCKET s)

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.168 MHA_TCP::Sockread_Event Class Reference 485

5.168 MHA_TCP::Sockread_Event Class Reference

Watch socket for incoming data.

Inheritance diagram for MHA_TCP::Sockread_Event:

MHA_TCP::Sockread_Event

MHA_TCP::Wakeup_Event

Public Member Functions

•Sockread_Event (SOCKET s)

Set socket to watch for.

Additional Inherited Members

5.168.1 Detailed Description

Watch socket for incoming data.

5.168.2 Constructor & Destructor Documentation

5.168.2.1 MHA_TCP::Sockread_Event::Sockread_Event (

SOCKET s)

Set socket to watch for.

Parameters

sThe socket to observe incoming data on.

The documentation for this class was generated from the following ﬁles:

486 CONTENTS

•mha_tcp.hh

•mha_tcp.cpp

5.169 MHA_TCP::Sockwrite_Event Class Reference

Inheritance diagram for MHA_TCP::Sockwrite_Event:

MHA_TCP::Sockwrite

_Event

MHA_TCP::Wakeup_Event

Public Member Functions

•Sockwrite_Event (SOCKET s)

Additional Inherited Members

5.169.1 Constructor & Destructor Documentation

5.169.1.1 MHA_TCP::Sockwrite_Event::Sockwrite_Event (

SOCKET s)

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.170 MHA_TCP::Thread Class Reference

A very simple class for portable threads.

5.170 MHA_TCP::Thread Class Reference 487

Public Types

• typedef void ∗(∗thr_f) (void ∗)

The thread function signature to use with this class.

Public Member Functions

•Thread (thr_f func, void ∗arg=0)

Constructor starts a new thread.

• virtual ∼Thread ()

The destructor should only be called when the Thread (p. 486) is ﬁnished.

• virtual void run ()

The internal method that delegated the new thread to the registered Thread (p. 486) function.

Public Attributes

•Async_Notify thread_ﬁnish_event

Event will be triggered when the thread exits.

• enum MHA_TCP::Thread:: { ... } state

The current state of the thread.

•thr_f thread_func

The thread function that the client has registered.

• void ∗thread_arg

The argument that the client wants to be handed through to the thread function.

•MHA_Error ∗error

The MHA_Error (p. 445) that caused the thread to abort, if any.

Protected Member Functions

•Thread ()

Default constructor may only be used by derived classes that want to start the thread them-

selves.

Protected Attributes

• void ∗arg

The argument for the client's thread function.

• void ∗return_value

The return value from the client's thread function is stored here When that function returns.

488 CONTENTS

Private Attributes

• pthread_t thread_handle

The posix thread handle.

• pthread_attr_t thread_attr

The posix thread attribute structure.

5.170.1 Detailed Description

A very simple class for portable threads.

5.170.2 Member Typedef Documentation

5.170.2.1 typedef void∗(∗MHA_TCP::Thread::thr_f) (void ∗)

The thread function signature to use with this class.

Derive from this class and call protected standard constructor to start threads differently.

5.170.3 Member Enumeration Documentation

5.170.3.1 anonymous enum

The current state of the thread.

Enumerator

PREPARED

RUNNING

FINISHED

5.170.4 Constructor & Destructor Documentation

5.170.4.1 MHA_TCP::Thread::Thread ( ) [protected]

Default constructor may only be used by derived classes that want to start the thread them-

selves.

5.170.4.2 Thread::Thread (

Thread::thr_f func,

void ∗arg = 0)

Constructor starts a new thread.

5.170 MHA_TCP::Thread Class Reference 489

Parameters

func The function to be executed by the thread.

arg The argument given to pass to the thread function.

5.170.4.3 Thread::∼Thread ( ) [virtual]

The destructor should only be called when the Thread (p. 486) is ﬁnished.

There is preliminary support for forceful thread cancellation in the destructor, but probably not

very robust or portable..

5.170.5 Member Function Documentation

5.170.5.1 void Thread::run ( ) [virtual]

The internal method that delegated the new thread to the registered Thread (p. 486) function.

5.170.6 Member Data Documentation

5.170.6.1 pthread_t MHA_TCP::Thread::thread_handle [private]

The posix thread handle.

5.170.6.2 pthread_attr_t MHA_TCP::Thread::thread_attr [private]

The posix thread attribute structure.

Required for starting a thread in detached state. Detachment is required to eliminate the need

for joining this thread.

5.170.6.3 void∗MHA_TCP::Thread::arg [protected]

The argument for the client's thread function.

5.170.6.4 void∗MHA_TCP::Thread::return_value [protected]

The return value from the client's thread function is stored here When that function returns.

5.170.6.5 Async_Notify MHA_TCP::Thread::thread_ﬁnish_event

Event will be triggered when the thread exits.

490 CONTENTS

5.170.6.6 enum { ... } MHA_TCP::Thread::state

The current state of the thread.

5.170.6.7 thr_f MHA_TCP::Thread::thread_func

The thread function that the client has registered.

5.170.6.8 void∗MHA_TCP::Thread::thread_arg

The argument that the client wants to be handed through to the thread function.

5.170.6.9 MHA_Error∗MHA_TCP::Thread::error

The MHA_Error (p. 445) that caused the thread to abort, if any.

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.171 MHA_TCP::Timeout_Event Class Reference

Inheritance diagram for MHA_TCP::Timeout_Event:

MHA_TCP::Timeout_Event

MHA_TCP::Wakeup_Event

Public Member Functions

•Timeout_Event (double interval)

• virtual OS_EVENT_TYPE get_os_event ()

5.172 MHA_TCP::Timeout_Watcher Class Reference 491

Private Attributes

• double end_time

Additional Inherited Members

5.171.1 Constructor & Destructor Documentation

5.171.1.1 Timeout_Event::Timeout_Event (

double interval )

5.171.2 Member Function Documentation

5.171.2.1 OS_EVENT_TYPE Timeout_Event::get_os_event ( ) [virtual]

Reimplemented from MHA_TCP::Wakeup_Event (p. 494).

5.171.3 Member Data Documentation

5.171.3.1 double MHA_TCP::Timeout_Event::end_time [private]

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.172 MHA_TCP::Timeout_Watcher Class Reference

OS-independent event watcher with internal ﬁxed-end-time timeout.

Inheritance diagram for MHA_TCP::Timeout_Watcher:

MHA_TCP::Timeout_Watcher

MHA_TCP::Event_Watcher

492 CONTENTS

Public Member Functions

•Timeout_Watcher (double interval)

• virtual ∼Timeout_Watcher ()

Private Attributes

•Timeout_Event timeout

Additional Inherited Members

5.172.1 Detailed Description

OS-independent event watcher with internal ﬁxed-end-time timeout.

5.172.2 Constructor & Destructor Documentation

5.172.2.1 Timeout_Watcher::Timeout_Watcher (

double interval )[explicit]

5.172.2.2 Timeout_Watcher::∼Timeout_Watcher ( ) [virtual]

5.172.3 Member Data Documentation

5.172.3.1 Timeout_Event MHA_TCP::Timeout_Watcher::timeout [private]

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.173 MHA_TCP::Wakeup_Event Class Reference 493

5.173 MHA_TCP::Wakeup_Event Class Reference

A base class for asynchronous wakeup events.

Inheritance diagram for MHA_TCP::Wakeup_Event:

MHA_TCP::Wakeup_Event

MHA_TCP::Async_Notify

MHA_TCP::Sockaccept

_Event

MHA_TCP::Sockread_Event

MHA_TCP::Sockwrite

_Event

MHA_TCP::Timeout_Event

Public Member Functions

•Wakeup_Event ()

Event Constructor.

• virtual void observed_by (Event_Watcher ∗observer)

Called by the Event_Watcher (p. 478) when this event is added to its list of observed events.

• virtual void ignored_by (Event_Watcher ∗observer)

Called by the Event_Watcher (p. 478) when this event is removed from its list of observed

events.

• virtual ∼Wakeup_Event ()

Destructor deregisters from observers.

• virtual OS_EVENT_TYPE get_os_event ()

Get necessary information for the Event Watcher.

• virtual void reset ()

For pure notiﬁcation events, reset the "signalled" status.

• virtual bool status ()

Query wether the event is in signalled state now.

Protected Attributes

•OS_EVENT_TYPE os_event

• bool os_event_valid

494 CONTENTS

Private Attributes

•std::set<class Event_Watcher ∗>observers

A list of all Event_Watcher (p. 478) instances that this Wakeup_Event (p. 493) is observed by

(stored here for proper deregistering).

5.173.1 Detailed Description

A base class for asynchronous wakeup events.

5.173.2 Constructor & Destructor Documentation

5.173.2.1 Wakeup_Event::Wakeup_Event ( )

Event Constructor.

The new event has invalid state.

5.173.2.2 Wakeup_Event::∼Wakeup_Event ( ) [virtual]

Destructor deregisters from observers.

5.173.3 Member Function Documentation

5.173.3.1 void Wakeup_Event::observed_by (

Event_Watcher ∗observer )[virtual]

Called by the Event_Watcher (p. 478) when this event is added to its list of observed events.

5.173.3.2 void Wakeup_Event::ignored_by (

Event_Watcher ∗observer )[virtual]

Called by the Event_Watcher (p. 478) when this event is removed from its list of observed

events.

5.173.3.3 OS_EVENT_TYPE Wakeup_Event::get_os_event ( ) [virtual]

Get necessary information for the Event Watcher.

Reimplemented in MHA_TCP::Timeout_Event (p. 491).

5.174 mha_tictoc_t Struct Reference 495

5.173.3.4 void Wakeup_Event::reset ( ) [virtual]

For pure notiﬁcation events, reset the "signalled" status.

Reimplemented in MHA_TCP::Async_Notify (p. 469).

5.173.3.5 bool Wakeup_Event::status ( ) [virtual]

Query wether the event is in signalled state now.

5.173.4 Member Data Documentation

5.173.4.1 std::set<class Event_Watcher ∗>MHA_TCP::Wakeup_Event::observers [private]

A list of all Event_Watcher (p. 478) instances that this Wakeup_Event (p. 493) is observed by

(stored here for proper deregistering).

5.173.4.2 OS_EVENT_TYPE MHA_TCP::Wakeup_Event::os_event [protected]

5.173.4.3 bool MHA_TCP::Wakeup_Event::os_event_valid [protected]

The documentation for this class was generated from the following ﬁles:

•mha_tcp.hh

•mha_tcp.cpp

5.174 mha_tictoc_t Struct Reference

Public Attributes

• struct timeval tv1

• struct timeval tv2

• struct timezone tz

• ﬂoat t

5.174.1 Member Data Documentation

5.174.1.1 struct timeval mha_tictoc_t::tv1

5.174.1.2 struct timeval mha_tictoc_t::tv2

5.174.1.3 struct timezone mha_tictoc_t::tz

5.174.1.4 ﬂoat mha_tictoc_t::t

The documentation for this struct was generated from the following ﬁle:

•mha_proﬁling.h

496 CONTENTS

5.175 mha_wave_t Struct Reference

Waveform signal structure.

Inheritance diagram for mha_wave_t:

mha_wave_t MHASignal::waveform_t

addsndfile::level_adapt_t

addsndfile::waveform

_proxy_t

fader_wave::level_adapt_t

gain::scaler_t

MHA_AC::waveform_t

MHAFilter::o1_ar_filter_t

MHAOvlFilter::fftfb_t

MHASignal::async_rmslevel_t

MHASignal::loop_wavefragment_t

MHASignal::ringbuffer_t

MHASignal::schroeder_t

MHASndFile::sf_wave_t

MHAWindow::base_t

speechnoise_t

addsndfile::resampled

_soundfile_t

MHAMultiSrc::waveform_t

mon_t

MHAFilter::o1flt_lowpass_t

MHAFilter::o1flt_maxtrack_t

MHAFilter::o1flt_mintrack_t

coherence::cohflt_t

fftfbpow::fftfbpow_t

MHAOvlFilter::overlap

_save_filterbank_t

multibandcompressor

::fftfb_plug_t

MHAOvlFilter::overlap

_save_filterbank_analytic_t fftfilterbank::fftfb

_plug_t

addsndfile::sndfile_t

MHAWindow::fun_t

MHAWindow::user_t

MHAWindow::bartlett_t

MHAWindow::blackman_t

MHAWindow::hamming_t

MHAWindow::hanning_t

MHAWindow::rect_t

Public Attributes

•mha_real_t ∗buf

signal buffer

• unsigned int num_channels

number of channels

• unsigned int num_frames

number of frames in each channel

•mha_channel_info_t ∗channel_info

detailed channel description

5.175.1 Detailed Description

Waveform signal structure.

This structure contains one fragment of a waveform signal. The member num_frames describes

the number of audio samples in each audio channel.

The ﬁeld channel_info must be an array of num_channels entries or NULL.

5.176 mhachain::chain_base_t Class Reference 497

5.175.2 Member Data Documentation

5.175.2.1 mha_real_t∗mha_wave_t::buf

signal buffer

5.175.2.2 unsigned int mha_wave_t::num_channels

number of channels

5.175.2.3 unsigned int mha_wave_t::num_frames

number of frames in each channel

5.175.2.4 mha_channel_info_t∗mha_wave_t::channel_info

detailed channel description

The documentation for this struct was generated from the following ﬁle:

•mha.hh

5.176 mhachain::chain_base_t Class Reference

Inheritance diagram for mhachain::chain_base_t:

mhachain::chain_base_t

mhachain::mhachain_t

MHAPlugin::plugin_t

< mhachain::plugs_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< mhachain::plugs_t >

498 CONTENTS

Public Member Functions

•chain_base_t (algo_comm_t, const std::string &, const std::string &)

• void process (mha_wave_t ∗,mha_wave_t ∗∗)

• void process (mha_spec_t ∗,mha_wave_t ∗∗)

• void process (mha_wave_t ∗,mha_spec_t ∗∗)

• void process (mha_spec_t ∗,mha_spec_t ∗∗)

• void prepare (mhaconﬁg_t &)

• void release ()

Protected Attributes

•MHAParser::bool_t bproﬁling

•MHAParser::vstring_t algos

Private Member Functions

• void update ()

Private Attributes

• std::vector<std::string >old_algos

•MHAEvents::patchbay_t<mhachain::chain_base_t >patchbay

•mhaconﬁg_t cﬁn

•mhaconﬁg_t cfout

• bool b_prepared

• std::string chain

Additional Inherited Members

5.176.1 Constructor & Destructor Documentation

5.176.1.1 mhachain::chain_base_t::chain_base_t (

algo_comm_t iac,

const std::string & ichain,

const std::string & ialgo )

5.176.2 Member Function Documentation

5.176.2.1 void mhachain::chain_base_t::process (

mha_wave_t ∗sin,

mha_wave_t ∗∗ sout )

5.176 mhachain::chain_base_t Class Reference 499

5.176.2.2 void mhachain::chain_base_t::process (

mha_spec_t ∗sin,

mha_wave_t ∗∗ sout )

5.176.2.3 void mhachain::chain_base_t::process (

mha_wave_t ∗sin,

mha_spec_t ∗∗ sout )

5.176.2.4 void mhachain::chain_base_t::process (

mha_spec_t ∗sin,

mha_spec_t ∗∗ sout )

5.176.2.5 void mhachain::chain_base_t::prepare (

mhaconﬁg_t &cf )[virtual]

Implements MHAPlugin::plugin_t<mhachain::plugs_t >(p. 730).

5.176.2.6 void mhachain::chain_base_t::release (

void ) [virtual]

Reimplemented from MHAPlugin::plugin_t<mhachain::plugs_t >(p. 731).

5.176.2.7 void mhachain::chain_base_t::update ( ) [private]

5.176.3 Member Data Documentation

5.176.3.1 MHAParser::bool_t mhachain::chain_base_t::bproﬁling [protected]

5.176.3.2 MHAParser::vstring_t mhachain::chain_base_t::algos [protected]

5.176.3.3 std::vector<std::string>mhachain::chain_base_t::old_algos [private]

5.176.3.4 MHAEvents::patchbay_t<mhachain::chain_base_t >

mhachain::chain_base_t::patchbay [private]

5.176.3.5 mhaconﬁg_t mhachain::chain_base_t::cﬁn [private]

5.176.3.6 mhaconﬁg_t mhachain::chain_base_t::cfout [private]

5.176.3.7 bool mhachain::chain_base_t::b_prepared [private]

5.176.3.8 std::string mhachain::chain_base_t::chain [private]

The documentation for this class was generated from the following ﬁles:

•mha_generic_chain.h

•mha_generic_chain.cpp

500 CONTENTS

5.177 mhachain::mhachain_t Class Reference

Inheritance diagram for mhachain::mhachain_t:

mhachain::mhachain_t

mhachain::chain_base_t

MHAPlugin::plugin_t

< mhachain::plugs_t >

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< mhachain::plugs_t >

Public Member Functions

•mhachain_t (algo_comm_t iac, const std::string &ichain, const std::string &ialgo)

Additional Inherited Members

5.177.1 Constructor & Destructor Documentation

5.177.1.1 mhachain::mhachain_t::mhachain_t (

algo_comm_t iac,

const std::string & ichain,

const std::string & ialgo )

The documentation for this class was generated from the following ﬁle:

•mhachain.cpp

5.178 mhachain::plugs_t Class Reference 501

5.178 mhachain::plugs_t Class Reference

Public Member Functions

•plugs_t (std::vector<std::string >algos,mhaconﬁg_t cﬁn, mhaconﬁg_t cfout, bool

do_prepare, MHAParser::parser_t &p, algo_comm_t iac, std::string ichain, bool use_←-

proﬁling)

•∼plugs_t ()

• void prepare (mhaconﬁg_t &)

• void release ()

• void process (mha_wave_t ∗,mha_spec_t ∗,mha_wave_t ∗∗,mha_spec_t ∗∗)

• bool prepared () const

Private Member Functions

• void alloc_plugs (std::vector<std::string >algos)

• void cleanup_plugs ()

• void update_proc_load ()

Private Attributes

• bool b_prepared

• std::vector<PluginLoader::mhapluginloader_t ∗>algos

•MHAParser::parser_t &parser

•algo_comm_t ac

• std::string chain

•MHAParser::parser_t proﬁling

•MHAParser::vstring_mon_t prof_algos

•MHAParser::vﬂoat_mon_t prof_init

•MHAParser::vﬂoat_mon_t prof_prepare

•MHAParser::vﬂoat_mon_t prof_release

•MHAParser::vﬂoat_mon_t prof_process

•MHAParser::ﬂoat_mon_t prof_process_tt

•MHAParser::vﬂoat_mon_t prof_process_load

• unsigned int proc_cnt

•mhaconﬁg_t prof_cfg

•MHAEvents::connector_t<mhachain::plugs_t >prof_load_con

•MHAEvents::connector_t<mhachain::plugs_t >prof_tt_con

• bool b_use_proﬁling

•mha_platform_tictoc_t tictoc

502 CONTENTS

5.178.1 Constructor & Destructor Documentation

5.178.1.1 mhachain::plugs_t::plugs_t (

std::vector<std::string >algos,

mhaconﬁg_t cﬁn,

mhaconﬁg_t cfout,

bool do_prepare,

MHAParser::parser_t &p,

algo_comm_t iac,

std::string ichain,

bool use_proﬁling )

5.178.1.2 mhachain::plugs_t::∼plugs_t ( )

5.178.2 Member Function Documentation

5.178.2.1 void mhachain::plugs_t::prepare (

mhaconﬁg_t &tf )

5.178.2.2 void mhachain::plugs_t::release (

void )

5.178.2.3 void mhachain::plugs_t::process (

mha_wave_t ∗win,

mha_spec_t ∗sin,

mha_wave_t ∗∗ wout,

mha_spec_t ∗∗ sout )

5.178.2.4 bool mhachain::plugs_t::prepared ( ) const [inline]

5.178.2.5 void mhachain::plugs_t::alloc_plugs (

std::vector<std::string >algos )[private]

5.178.2.6 void mhachain::plugs_t::cleanup_plugs ( ) [private]

5.178.2.7 void mhachain::plugs_t::update_proc_load ( ) [private]

5.178.3 Member Data Documentation

5.178.3.1 bool mhachain::plugs_t::b_prepared [private]

5.178.3.2 std::vector<PluginLoader::mhapluginloader_t∗>mhachain::plugs_t::algos

[private]

5.178.3.3 MHAParser::parser_t& mhachain::plugs_t::parser [private]

5.178 mhachain::plugs_t Class Reference 503

5.178.3.4 algo_comm_t mhachain::plugs_t::ac [private]

5.178.3.5 std::string mhachain::plugs_t::chain [private]

5.178.3.6 MHAParser::parser_t mhachain::plugs_t::proﬁling [private]

5.178.3.7 MHAParser::vstring_mon_t mhachain::plugs_t::prof_algos [private]

5.178.3.8 MHAParser::vﬂoat_mon_t mhachain::plugs_t::prof_init [private]

5.178.3.9 MHAParser::vﬂoat_mon_t mhachain::plugs_t::prof_prepare [private]

5.178.3.10 MHAParser::vﬂoat_mon_t mhachain::plugs_t::prof_release [private]

5.178.3.11 MHAParser::vﬂoat_mon_t mhachain::plugs_t::prof_process [private]

5.178.3.12 MHAParser::ﬂoat_mon_t mhachain::plugs_t::prof_process_tt [private]

5.178.3.13 MHAParser::vﬂoat_mon_t mhachain::plugs_t::prof_process_load [private]

5.178.3.14 unsigned int mhachain::plugs_t::proc_cnt [private]

5.178.3.15 mhaconﬁg_t mhachain::plugs_t::prof_cfg [private]

5.178.3.16 MHAEvents::connector_t<mhachain::plugs_t>mhachain::plugs_t::prof_load_con

[private]

5.178.3.17 MHAEvents::connector_t<mhachain::plugs_t>mhachain::plugs_t::prof_tt_con

[private]

5.178.3.18 bool mhachain::plugs_t::b_use_proﬁling [private]

5.178.3.19 mha_platform_tictoc_t mhachain::plugs_t::tictoc [private]

The documentation for this class was generated from the following ﬁles:

•mha_generic_chain.h

•mha_generic_chain.cpp

504 CONTENTS

5.179 mhaconﬁg_t Struct Reference

MHA prepare conﬁguration structure.

Inheritance diagram for mhaconﬁg_t:

mhaconfig_t

coherence::cohflt_t

Public Attributes

• unsigned int channels

Number of audio channels.

• unsigned int domain

Signal domain (MHA_WAVEFORM or MHA_SPECTRUM)

• unsigned int fragsize

Fragment size of waveform data.

• unsigned int wndlen

Window length of spectral data.

• unsigned int fftlen

FFT length of spectral data.

•mha_real_t srate

Sampling rate in Hz.

5.179.1 Detailed Description

MHA prepare conﬁguration structure.

This structure contains information about channel number and domain for input and output

signals of a openMHA Plugin. Each plugin can change any of these parameters, e.g. by

resampling of the signal. The only limitation is that the callback frequency is ﬁxed (except for

the plugins db and dbasync).

Todo Add information on number of bands and on center frequencies, or replace by mha_←-

audio_descriptor_t (p. 427).

5.179 mhaconﬁg_t Struct Reference 505

5.179.2 Member Data Documentation

5.179.2.1 unsigned int mhaconﬁg_t::channels

Number of audio channels.

5.179.2.2 unsigned int mhaconﬁg_t::domain

Signal domain (MHA_WAVEFORM or MHA_SPECTRUM)

5.179.2.3 unsigned int mhaconﬁg_t::fragsize

Fragment size of waveform data.

5.179.2.4 unsigned int mhaconﬁg_t::wndlen

Window length of spectral data.

5.179.2.5 unsigned int mhaconﬁg_t::fftlen

FFT length of spectral data.

5.179.2.6 mha_real_t mhaconﬁg_t::srate

Sampling rate in Hz.

The documentation for this struct was generated from the following ﬁle:

•mha.hh

506 CONTENTS

5.180 MHAEvents::connector_base_t Class Reference

Inheritance diagram for MHAEvents::connector_base_t:

MHAEvents::connector

_base_t

MHAEvents::connector

_t< mhachain::mhachain

::plugs_t >

MHAEvents::connector

_t< MHAOvlFilter::fscale

_bw_t >

MHAEvents::connector

_t< MHAOvlFilter::fscale_t >

MHAEvents::connector

_t< MHAParser::mhapluginloader_t >

MHAEvents::connector

_t< receiver_t >

Public Member Functions

•connector_base_t ()

• virtual ∼connector_base_t ()

• virtual void emit_event ()

• virtual void emit_event (const std::string &)

• virtual void emit_event (const std::string &, unsigned int, unsigned int)

• void emitter_die ()

Protected Attributes

• bool emitter_is_alive

5.180.1 Constructor & Destructor Documentation

5.180.1.1 MHAEvents::connector_base_t::connector_base_t ( )

5.180.1.2 MHAEvents::connector_base_t::∼connector_base_t ( ) [virtual]

5.180 MHAEvents::connector_base_t Class Reference 507

5.180.2 Member Function Documentation

5.180.2.1 void MHAEvents::connector_base_t::emit_event ( ) [virtual]

Reimplemented in MHAEvents::connector_t<receiver_t >(p. 509), MHAEvents←-

::connector_t<MHAOvlFilter::fscale_bw_t >(p. 509), MHAEvents::connector_t<

MHAParser::mhapluginloader_t >(p. 509), MHAEvents::connector_t<mhachain←-

::mhachain::plugs_t >(p. 509), and MHAEvents::connector_t<MHAOvlFilter::fscale_t >

(p. 509).

5.180.2.2 void MHAEvents::connector_base_t::emit_event (

const std::string & ) [virtual]

Reimplemented in MHAEvents::connector_t<receiver_t >(p. 509), MHAEvents←-

::connector_t<MHAOvlFilter::fscale_bw_t >(p. 509), MHAEvents::connector_t<

MHAParser::mhapluginloader_t >(p. 509), MHAEvents::connector_t<mhachain←-

::mhachain::plugs_t >(p. 509), and MHAEvents::connector_t<MHAOvlFilter::fscale_t >

(p. 509).

5.180.2.3 void MHAEvents::connector_base_t::emit_event (

const std::string & ,

unsigned int ,

unsigned int ) [virtual]

Reimplemented in MHAEvents::connector_t<receiver_t >(p. 509), MHAEvents←-

::connector_t<MHAOvlFilter::fscale_bw_t >(p. 509), MHAEvents::connector_t<

MHAParser::mhapluginloader_t >(p. 509), MHAEvents::connector_t<mhachain←-

::mhachain::plugs_t >(p. 509), and MHAEvents::connector_t<MHAOvlFilter::fscale_t >

(p. 509).

5.180.2.4 void MHAEvents::connector_base_t::emitter_die ( )

5.180.3 Member Data Documentation

5.180.3.1 bool MHAEvents::connector_base_t::emitter_is_alive [protected]

The documentation for this class was generated from the following ﬁles:

•mha_event_emitter.h

•mha_events.cpp

508 CONTENTS

5.181 MHAEvents::connector_t<receiver_t >Class Template Reference

Inheritance diagram for MHAEvents::connector_t<receiver_t >:

MHAEvents::connector

_t< receiver_t >

MHAEvents::connector

_base_t

Public Member Functions

•connector_t (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)())

•connector_t (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)(const std::string &))

•connector_t (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)(const std::string &, unsigned

int, unsigned int))

•∼connector_t ()

Private Member Functions

• void emit_event ()

• void emit_event (const std::string &)

• void emit_event (const std::string &, unsigned int, unsigned int)

Private Attributes

•emitter_t ∗emitter

• receiver_t ∗receiver

• void(receiver_t::∗eventhandler )()

• void(receiver_t::∗eventhandler_s )(const std::string &)

• void(receiver_t::∗eventhandler_suu )(const std::string &, unsigned int, unsigned int)

5.181 MHAEvents::connector_t<receiver_t >Class Template

Reference 509

Additional Inherited Members

5.181.1 Constructor & Destructor Documentation

5.181.1.1 template<class receiver_t>MHAEvents::connector_t<receiver_t >::connector_t (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)() rfun )

5.181.1.2 template<class receiver_t>MHAEvents::connector_t<receiver_t >::connector_t (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)(const std::string &) rfun )

5.181.1.3 template<class receiver_t>MHAEvents::connector_t<receiver_t >::connector_t (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)(const std::string &, unsigned int, unsigned int) rfun )

5.181.1.4 template<class receiver_t >MHAEvents::connector_t<receiver_t >::∼connector_t (

)

5.181.2 Member Function Documentation

5.181.2.1 template<class receiver_t >void MHAEvents::connector_t<receiver_t >::emit_event (

)[private],[virtual]

Reimplemented from MHAEvents::connector_base_t (p. 507).

5.181.2.2 template<class receiver_t >void MHAEvents::connector_t<receiver_t >::emit_event (

const std::string & arg )[private],[virtual]

Reimplemented from MHAEvents::connector_base_t (p. 507).

5.181.2.3 template<class receiver_t >void MHAEvents::connector_t<receiver_t >::emit_event (

const std::string & arg,

unsigned int arg2,

unsigned int arg3 )[private],[virtual]

Reimplemented from MHAEvents::connector_base_t (p. 507).

510 CONTENTS

5.181.3 Member Data Documentation

5.181.3.1 template<class receiver_t>emitter_t∗MHAEvents::connector_t<receiver_t

>::emitter [private]

5.181.3.2 template<class receiver_t>receiver_t∗MHAEvents::connector_t<receiver_t

>::receiver [private]

5.181.3.3 template<class receiver_t>void(receiver_t::∗MHAEvents::connector_t<receiver_t

>::eventhandler) () [private]

5.181.3.4 template<class receiver_t>void(receiver_t::∗MHAEvents::connector_t<receiver_t

>::eventhandler_s) (const std::string &) [private]

5.181.3.5 template<class receiver_t>void(receiver_t::∗MHAEvents::connector_t<receiver_t

>::eventhandler_suu) (const std::string &, unsigned int, unsigned int) [private]

The documentation for this class was generated from the following ﬁle:

•mha_events.h

5.182 MHAEvents::emitter_t Class Reference

Class for emitting openMHA events.

Public Member Functions

•∼emitter_t ()

• void operator() ()

Emit an event without parameter.

• void operator() (const std::string &)

Emit an event with string parameter.

• void operator() (const std::string &, unsigned int, unsigned int)

Emit an event with string parameter and two unsigned int parameters.

• void connect (connector_base_t ∗)

• void disconnect (connector_base_t ∗)

Private Attributes

• std::list<connector_base_t ∗>connections

5.182 MHAEvents::emitter_t Class Reference 511

5.182.1 Detailed Description

Class for emitting openMHA events.

Use the template claas MHAEvents::patchbay_t (p. 512) for connecting to an emitter.

5.182.2 Constructor & Destructor Documentation

5.182.2.1 MHAEvents::emitter_t::∼emitter_t ( )

5.182.3 Member Function Documentation

5.182.3.1 void MHAEvents::emitter_t::operator() ( )

Emit an event without parameter.

5.182.3.2 void MHAEvents::emitter_t::operator() (

const std::string & arg )

Emit an event with string parameter.

5.182.3.3 void MHAEvents::emitter_t::operator() (

const std::string & arg,

unsigned int arg2,

unsigned int arg3 )

Emit an event with string parameter and two unsigned int parameters.

5.182.3.4 void MHAEvents::emitter_t::connect (

connector_base_t ∗c)

5.182.3.5 void MHAEvents::emitter_t::disconnect (

connector_base_t ∗c)

5.182.4 Member Data Documentation

5.182.4.1 std::list<connector_base_t∗>MHAEvents::emitter_t::connections [private]

The documentation for this class was generated from the following ﬁles:

•mha_event_emitter.h

•mha_events.cpp

512 CONTENTS

5.183 MHAEvents::patchbay_t<receiver_t >Class Template Reference

Patchbay which connects any event emitter with any member function of the parameter class.

Public Member Functions

•∼patchbay_t ()

• void connect (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)())

Connect a receiver member function void (receiver_t::∗)() with an event emitter.

• void connect (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)(const std::string &))

Connect a receiver member function void (receiver_t::∗)(const std::string&) with an event emit-

ter.

• void connect (emitter_t ∗, receiver_t ∗, void(receiver_t::∗)(const std::string &, unsigned

int, unsigned int))

Private Attributes

• std::list<connector_t<receiver_t >∗>cons

5.183.1 Detailed Description

template<class receiver_t>

class MHAEvents::patchbay_t<receiver_t >

Patchbay which connects any event emitter with any member function of the parameter class.

The connections created by the connect() (p. 512) function are hold until the destructor is

called. To avoid access to invalid function pointers, it is required to destruct the patchbay

before the receiver, usually by declaring the patchbay as a member of the receiver.

The receiver can be any claas or structure; the event callback can be either a member function

without arguments or with const std::string& argument.

5.183.2 Constructor & Destructor Documentation

5.183.2.1 template<class receiver_t >MHAEvents::patchbay_t<receiver_t >::∼patchbay_t ( )

5.183.3 Member Function Documentation

5.183.3.1 template<class receiver_t>void MHAEvents::patchbay_t<receiver_t >::connect (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)() rfun )

Connect a receiver member function void (receiver_t::∗)() with an event emitter.

Create a connection.

The connection is removed when the patchbay is destructed.

5.184 MHAFilter::adapt_ﬁlter_param_t Class Reference 513

Parameters

ePointer to an event emitter

rPointer to the receiver

rfun Pointer to a member function of the receiver class

5.183.3.2 template<class receiver_t>void MHAEvents::patchbay_t<receiver_t >::connect (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)(const std::string &) rfun )

Connect a receiver member function void (receiver_t::∗)(const std::string&) with an event emit-

ter.

Create a connection.

The connection is removed when the patchbay is destructed.

Parameters

ePointer to an event emitter

rPointer to the receiver

rfun Pointer to a member function of the receiver class

5.183.3.3 template<class receiver_t>void MHAEvents::patchbay_t<receiver_t >::connect (

emitter_t ∗e,

receiver_t ∗r,

void(receiver_t::∗)(const std::string &, unsigned int, unsigned int) rfun )

5.183.4 Member Data Documentation

5.183.4.1 template<class receiver_t>std::list<connector_t<receiver_t>∗>

MHAEvents::patchbay_t<receiver_t >::cons [private]

The documentation for this class was generated from the following ﬁle:

•mha_events.h

5.184 MHAFilter::adapt_ﬁlter_param_t Class Reference

Public Member Functions

•adapt_ﬁlter_param_t (mha_real_t imu, bool ierr_in)

514 CONTENTS

Public Attributes

•mha_real_t mu

• bool err_in

5.184.1 Constructor & Destructor Documentation

5.184.1.1 MHAFilter::adapt_ﬁlter_param_t::adapt_ﬁlter_param_t (

mha_real_t imu,

bool ierr_in )

5.184.2 Member Data Documentation

5.184.2.1 mha_real_t MHAFilter::adapt_ﬁlter_param_t::mu

5.184.2.2 bool MHAFilter::adapt_ﬁlter_param_t::err_in

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.185 MHAFilter::adapt_ﬁlter_state_t Class Reference

Public Member Functions

•adapt_ﬁlter_state_t (int ntaps, int nchannels)

• void ﬁlter (mha_wave_t y, mha_wave_t e, mha_wave_t x, mha_wave_t d, mha_real←-

_t mu, bool err_in)

Private Attributes

• int ntaps

• int nchannels

•MHASignal::waveform_t W

•MHASignal::waveform_t X

•MHASignal::waveform_t od

•MHASignal::waveform_t oy

5.186 MHAFilter::adapt_ﬁlter_t Class Reference 515

5.185.1 Constructor & Destructor Documentation

5.185.1.1 MHAFilter::adapt_ﬁlter_state_t::adapt_ﬁlter_state_t (

int ntaps,

int nchannels )

5.185.2 Member Function Documentation

5.185.2.1 void MHAFilter::adapt_ﬁlter_state_t::ﬁlter (

mha_wave_t y,

mha_wave_t e,

mha_wave_t x,

mha_wave_t d,

mha_real_t mu,

bool err_in )

5.185.3 Member Data Documentation

5.185.3.1 int MHAFilter::adapt_ﬁlter_state_t::ntaps [private]

5.185.3.2 int MHAFilter::adapt_ﬁlter_state_t::nchannels [private]

5.185.3.3 MHASignal::waveform_t MHAFilter::adapt_ﬁlter_state_t::W [private]

5.185.3.4 MHASignal::waveform_t MHAFilter::adapt_ﬁlter_state_t::X [private]

5.185.3.5 MHASignal::waveform_t MHAFilter::adapt_ﬁlter_state_t::od [private]

5.185.3.6 MHASignal::waveform_t MHAFilter::adapt_ﬁlter_state_t::oy [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.186 MHAFilter::adapt_ﬁlter_t Class Reference

Adaptive ﬁlter.

Inheritance diagram for MHAFilter::adapt_ﬁlter_t:

MHAFilter::adapt_filter_t

MHAParser::parser_tMHAParser::base_t

MHAPlugin::config_t

< adapt_filter_state_t >

MHAPlugin::config_t

< adapt_filter_param_t >

516 CONTENTS

Public Member Functions

•adapt_ﬁlter_t (std::string)

• void ﬁlter (mha_wave_t y, mha_wave_t e, mha_wave_t x, mha_wave_t d)

• void set_channelcnt (unsigned int)

Private Member Functions

• void update_mu ()

• void update_ntaps ()

Private Attributes

•MHAParser::ﬂoat_t mu

•MHAParser::int_t ntaps

•MHAParser::bool_t err_in

•MHAEvents::patchbay_t<adapt_ﬁlter_t >connector

• unsigned int nchannels

Additional Inherited Members

5.186.1 Detailed Description

Adaptive ﬁlter.

5.186.2 Constructor & Destructor Documentation

5.186.2.1 MHAFilter::adapt_ﬁlter_t::adapt_ﬁlter_t (

std::string help )

5.186.3 Member Function Documentation

5.186.3.1 void MHAFilter::adapt_ﬁlter_t::ﬁlter (

mha_wave_t y,

mha_wave_t e,

mha_wave_t x,

mha_wave_t d)

5.186.3.2 void MHAFilter::adapt_ﬁlter_t::set_channelcnt (

unsigned int nch )

5.186.3.3 void MHAFilter::adapt_ﬁlter_t::update_mu ( ) [private]

5.187 MHAFilter::blockprocessing_polyphase_resampling_t Class

Reference 517

5.186.3.4 void MHAFilter::adapt_ﬁlter_t::update_ntaps ( ) [private]

5.186.4 Member Data Documentation

5.186.4.1 MHAParser::ﬂoat_t MHAFilter::adapt_ﬁlter_t::mu [private]

5.186.4.2 MHAParser::int_t MHAFilter::adapt_ﬁlter_t::ntaps [private]

5.186.4.3 MHAParser::bool_t MHAFilter::adapt_ﬁlter_t::err_in [private]

5.186.4.4 MHAEvents::patchbay_t<adapt_ﬁlter_t>MHAFilter::adapt_ﬁlter_t::connector

[private]

5.186.4.5 unsigned int MHAFilter::adapt_ﬁlter_t::nchannels [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.187 MHAFilter::blockprocessing_polyphase_resampling_t Class Reference

A class that does polyphase resampling and takes into account block processing.

Public Member Functions

•blockprocessing_polyphase_resampling_t (ﬂoat source_srate, unsigned source_←-

fragsize, ﬂoat target_srate, unsigned target_fragsize, ﬂoat nyquist_ratio, ﬂoat irslen, un-

signed nchannels, bool add_delay)

Contructs a polyphase resampling ﬁlter that can be used for blockprocessing with the given

parameters.

• virtual ∼blockprocessing_polyphase_resampling_t ()

• void write (mha_wave_t &signal)

Write signal to the ringbuffer.

• void read (mha_wave_t &signal)

Read resampled signal.

• bool can_read () const

Checks if the resampling ring buffer can produce another output signal block.

Private Attributes

•polyphase_resampling_t ∗resampling

• unsigned fragsize_in

• unsigned fragsize_out

• unsigned num_channels

518 CONTENTS

5.187.1 Detailed Description

A class that does polyphase resampling and takes into account block processing.

5.187.2 Constructor & Destructor Documentation

5.187.2.1 MHAFilter::blockprocessing_polyphase_resampling_t::blockprocessing_polyphase_←-

resampling_t (

ﬂoat source_srate,

unsigned source_fragsize,

ﬂoat target_srate,

unsigned target_fragsize,

ﬂoat nyquist_ratio,

ﬂoat irslen,

unsigned nchannels,

bool add_delay )

Contructs a polyphase resampling ﬁlter that can be used for blockprocessing with the given

parameters.

Parameters

source_srate Source sampling rate / Hz

source_fragsize Fragment size of incoming audio blocks / frames at source_srate

target_srate Target sampling rate / Hz

target_fragsize Fragment size of produced audio blocks / frames at target_srate

nyquist_ratio Low pass ﬁlter cutoff frequency relative to the nyquist frequency of the

smaller of the two sampling rates. Example values: 0.8, 0.9

irslen Impulse response length used for low pass ﬁltering / s

nchannels Number of audio channels

add_delay To avoid underruns, a delay is generally necessary for round trip block

size adaptations. It is only necessary to add this delay to one of the two

resampling chains. Set this parameter to true for the ﬁrst resampling

object of a round trip pair. It will add the necessary delay, and calculate

the size of the ring buffer appropriately, When set to false, only the

ringbuffer size will be set sufﬁciently.

5.187.2.2 virtual MHAFilter::blockprocessing_polyphase_resampling_t::∼blockprocessing_polyphase←-

_resampling_t ( ) [inline],[virtual]

5.187.3 Member Function Documentation

5.187.3.1 void MHAFilter::blockprocessing_polyphase_resampling_t::write (

mha_wave_t &signal )

Write signal to the ringbuffer.

5.187 MHAFilter::blockprocessing_polyphase_resampling_t Class

Reference 519

Parameters

signal input signal in original sampling rate

Exceptions

MHA_Error (p. 445)Raises exception if there is not enough room, if the number of

channels does not match, or if the number of frames is not equal to

the number speciﬁed in the constructor

5.187.3.2 void MHAFilter::blockprocessing_polyphase_resampling_t::read (

mha_wave_t &signal )

Read resampled signal.

Will perform the resampling and remove no longer needed samples from the input buffer.

Parameters

signal buffer to write the resampled signal to.

Exceptions

MHA_Error (p. 445)Raises exception if there is not enough input signal, if the number of

channels of frames does not match.

5.187.3.3 bool MHAFilter::blockprocessing_polyphase_resampling_t::can_read ( ) const [inline]

Checks if the resampling ring buffer can produce another output signal block.

5.187.4 Member Data Documentation

5.187.4.1 polyphase_resampling_t∗MHAFilter::blockprocessing_polyphase_resampling_t←-

::resampling [private]

5.187.4.2 unsigned MHAFilter::blockprocessing_polyphase_resampling_t::fragsize_in [private]

5.187.4.3 unsigned MHAFilter::blockprocessing_polyphase_resampling_t::fragsize_out [private]

5.187.4.4 unsigned MHAFilter::blockprocessing_polyphase_resampling_t::num_channels

[private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

520 CONTENTS

5.188 MHAFilter::complex_bandpass_t Class Reference

Complex bandpass ﬁlter.

Public Member Functions

•complex_bandpass_t (std::vector<mha_complex_t >A, std::vector<mha_←-

complex_t >B)

Constructor with ﬁlter coefﬁcients (one per channel)

• void set_state (mha_real_t val)

• void set_state (std::vector<mha_real_t >val)

• void set_state (mha_complex_t val)

• void set_weights (std::vector<mha_complex_t >new_B)

Allow to modify the input weights at a later stage.

• std::vector<mha_complex_t >get_weights () const

• void ﬁlter (const mha_wave_t &X, mha_spec_t &Y)

Filter method for real value input.

• void ﬁlter (const mha_wave_t &X, mha_wave_t &Yre, mha_wave_t &Yim)

Filter method for real value input.

• void ﬁlter (const mha_spec_t &X, mha_spec_t &Y)

Filter method for complex value input.

• void ﬁlter (const mha_wave_t &Xre, const mha_wave_t &Xim, mha_wave_t &Yre,

mha_wave_t &Yim)

Filter method for complex value input.

• std::string inspect () const

Static Public Member Functions

• static std::vector<mha_complex_t >creator_A (std::vector<mha_real_t >cf, std←-

::vector<mha_real_t >bw, mha_real_t srate, unsigned int order)

• static std::vector<mha_complex_t >creator_B (std::vector<mha_complex_t >A,

unsigned int order)

Private Attributes

• std::vector<mha_complex_t >A_

• std::vector<mha_complex_t >B_

• std::vector<mha_complex_t >Yn

5.188.1 Detailed Description

Complex bandpass ﬁlter.

5.188 MHAFilter::complex_bandpass_t Class Reference 521

5.188.2 Constructor & Destructor Documentation

5.188.2.1 MHAFilter::complex_bandpass_t::complex_bandpass_t (

std::vector<mha_complex_t >A,

std::vector<mha_complex_t >B)

Constructor with ﬁlter coefﬁcients (one per channel)

Parameters

Acomplex ﬁlter coefﬁcients, one per band

Bcomplex weights

5.188.3 Member Function Documentation

5.188.3.1 std::vector<mha_complex_t >MHAFilter::complex_bandpass_t::creator_A (

std::vector<mha_real_t >cf,

std::vector<mha_real_t >bw,

mha_real_t srate,

unsigned int order )[static]

5.188.3.2 std::vector<mha_complex_t >MHAFilter::complex_bandpass_t::creator_B (

std::vector<mha_complex_t >A,

unsigned int order )[static]

5.188.3.3 void MHAFilter::complex_bandpass_t::set_state (

mha_real_t val )

5.188.3.4 void MHAFilter::complex_bandpass_t::set_state (

std::vector<mha_real_t >val )

5.188.3.5 void MHAFilter::complex_bandpass_t::set_state (

mha_complex_t val )

5.188.3.6 void MHAFilter::complex_bandpass_t::set_weights (

std::vector<mha_complex_t >new_B )

Allow to modify the input weights at a later stage.

5.188.3.7 std::vector<mha_complex_t>MHAFilter::complex_bandpass_t::get_weights ( ) const

[inline]

5.188.3.8 void MHAFilter::complex_bandpass_t::ﬁlter (

const mha_wave_t &X,

mha_spec_t &Y)[inline]

Filter method for real value input.

522 CONTENTS

5.188.3.9 void MHAFilter::complex_bandpass_t::ﬁlter (

const mha_wave_t &X,

mha_wave_t &Yre,

mha_wave_t &Yim )[inline]

Filter method for real value input.

5.188.3.10 void MHAFilter::complex_bandpass_t::ﬁlter (

const mha_spec_t &X,

mha_spec_t &Y)[inline]

Filter method for complex value input.

5.188.3.11 void MHAFilter::complex_bandpass_t::ﬁlter (

const mha_wave_t &Xre,

const mha_wave_t &Xim,

mha_wave_t &Yre,

mha_wave_t &Yim )[inline]

Filter method for complex value input.

5.188.3.12 std::string MHAFilter::complex_bandpass_t::inspect ( ) const [inline]

5.188.4 Member Data Documentation

5.188.4.1 std::vector<mha_complex_t>MHAFilter::complex_bandpass_t::A_ [private]

5.188.4.2 std::vector<mha_complex_t>MHAFilter::complex_bandpass_t::B_ [private]

5.188.4.3 std::vector<mha_complex_t>MHAFilter::complex_bandpass_t::Yn [private]

The documentation for this class was generated from the following ﬁles:

•complex_ﬁlter.h

•complex_ﬁlter.cpp

5.189 MHAFilter::diff_t Class Reference

Differentiator class (non-normalized)

Inheritance diagram for MHAFilter::diff_t:

MHAFilter::diff_t

MHAFilter::filter_t

5.190 MHAFilter::fftﬁlter_t Class Reference 523

Public Member Functions

•diff_t (unsigned int ch)

Additional Inherited Members

5.189.1 Detailed Description

Differentiator class (non-normalized)

5.189.2 Constructor & Destructor Documentation

5.189.2.1 MHAFilter::diff_t::diff_t (

unsigned int ch )

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.190 MHAFilter::fftﬁlter_t Class Reference

FFT based FIR ﬁlter implementation.

Inheritance diagram for MHAFilter::fftﬁlter_t:

MHAFilter::fftfilter_t

MHAFilter::resampling

_filter_t

524 CONTENTS

Public Member Functions

•fftﬁlter_t (unsigned int fragsize, unsigned int channels, unsigned int fftlen)

Constructor.

•∼fftﬁlter_t ()

• void update_coeffs (const mha_wave_t ∗pwIRS)

Update the set of coefﬁcients.

• void ﬁlter (const mha_wave_t ∗pwIn, mha_wave_t ∗∗ppwOut, const mha_wave_t ∗pw←-

IRS)

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

• void ﬁlter (const mha_wave_t ∗pwIn, mha_wave_t ∗∗ppwOut)

Apply ﬁlter to waveform fragment, without changing the coefﬁcients.

• void ﬁlter (const mha_wave_t ∗pwIn, mha_wave_t ∗∗ppwOut, const mha_spec_t ∗ps←-

Weights)

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

Private Attributes

• unsigned int fragsize

• unsigned int channels

• unsigned int fftlen

•MHASignal::waveform_t wInput_fft

•mha_wave_t wInput

•MHASignal::waveform_t wOutput_fft

•mha_wave_t wOutput

•MHASignal::spectrum_t sInput

•MHASignal::spectrum_t sWeights

•MHASignal::waveform_t wIRS_fft

•mha_fft_t fft

5.190.1 Detailed Description

FFT based FIR ﬁlter implementation.

The maximal number of coefﬁcients can be FFT length - fragsize + 1.

5.190.2 Constructor & Destructor Documentation

5.190.2.1 MHAFilter::fftﬁlter_t::fftﬁlter_t (

unsigned int fragsize,

unsigned int channels,

unsigned int fftlen )

Constructor.

5.190 MHAFilter::fftﬁlter_t Class Reference 525

Parameters

fragsize Number of frames expected in input signal (each cycle).

channels Number of channels expected in input signal.

fftlen FFT length of ﬁlter.

5.190.2.2 MHAFilter::fftﬁlter_t::∼fftﬁlter_t ( )

5.190.3 Member Function Documentation

5.190.3.1 void MHAFilter::fftﬁlter_t::update_coeffs (

const mha_wave_t ∗pwIRS )

Update the set of coefﬁcients.

Parameters

pwIRS Coefﬁcients structure

Note

The number of channels in h must match the number of channels given in the constructor.

The ﬁlter length is limited to fftlen-fragsize+1 (longer IRS will be shortened).

5.190.3.2 void MHAFilter::fftﬁlter_t::ﬁlter (

const mha_wave_t ∗pwIn,

mha_wave_t ∗∗ ppwOut,

const mha_wave_t ∗pwIRS )

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

Parameters

pw←-

Input signal pointer.

Return values

ppwOut Pointer to output signal pointer, will be set to a valid signal.

Parameters

pwIRS Pointer to FIR coefﬁcients structure.

526 CONTENTS

5.190.3.3 void MHAFilter::fftﬁlter_t::ﬁlter (

const mha_wave_t ∗pwIn,

mha_wave_t ∗∗ ppwOut )

Apply ﬁlter to waveform fragment, without changing the coefﬁcients.

Parameters

pw←-

Input signal pointer.

Return values

ppwOut Pointer to output signal pointer, will be set to a valid signal

5.190.3.4 void MHAFilter::fftﬁlter_t::ﬁlter (

const mha_wave_t ∗pwIn,

mha_wave_t ∗∗ ppwOut,

const mha_spec_t ∗psWeights )

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

Parameters

pw←-

Input signal pointer.

Return values

ppwOut Pointer to output signal pointer, will be set to a valid signal.

Parameters

psWeights Pointer to ﬁlter weights structure.

5.190.4 Member Data Documentation

5.190.4.1 unsigned int MHAFilter::fftﬁlter_t::fragsize [private]

5.190.4.2 unsigned int MHAFilter::fftﬁlter_t::channels [private]

5.190.4.3 unsigned int MHAFilter::fftﬁlter_t::fftlen [private]

5.191 MHAFilter::fftﬁlterbank_t Class Reference 527

5.190.4.4 MHASignal::waveform_t MHAFilter::fftﬁlter_t::wInput_fft [private]

5.190.4.5 mha_wave_t MHAFilter::fftﬁlter_t::wInput [private]

5.190.4.6 MHASignal::waveform_t MHAFilter::fftﬁlter_t::wOutput_fft [private]

5.190.4.7 mha_wave_t MHAFilter::fftﬁlter_t::wOutput [private]

5.190.4.8 MHASignal::spectrum_t MHAFilter::fftﬁlter_t::sInput [private]

5.190.4.9 MHASignal::spectrum_t MHAFilter::fftﬁlter_t::sWeights [private]

5.190.4.10 MHASignal::waveform_t MHAFilter::fftﬁlter_t::wIRS_fft [private]

5.190.4.11 mha_fft_t MHAFilter::fftﬁlter_t::fft [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.191 MHAFilter::fftﬁlterbank_t Class Reference

FFT based FIR ﬁlterbank implementation.

Inheritance diagram for MHAFilter::fftﬁlterbank_t:

MHAFilter::fftfilterbank_t

MHAOvlFilter::overlap

_save_filterbank_t

MHAOvlFilter::overlap

_save_filterbank_analytic_t

fftfilterbank::fftfb

_plug_t

528 CONTENTS

Public Member Functions

•fftﬁlterbank_t (unsigned int fragsize, unsigned int inputchannels, unsigned int ﬁrchan-

nels, unsigned int fftlen)

Constructor.

•∼fftﬁlterbank_t ()

• void update_coeffs (const mha_wave_t ∗h)

Update the set of coefﬁcients.

• void ﬁlter (const mha_wave_t ∗s_in, mha_wave_t ∗∗s_out, const mha_wave_t ∗h)

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

• void ﬁlter (const mha_wave_t ∗s_in, mha_wave_t ∗∗s_out)

Apply ﬁlter to waveform fragment, without changing the coefﬁcients.

• const mha_wave_t ∗get_irs () const

Return the current IRS.

Private Attributes

• unsigned int fragsize

• unsigned int inputchannels

• unsigned int ﬁrchannels

• unsigned int outputchannels

• unsigned int fftlen

•MHASignal::waveform_t hw

•MHASignal::spectrum_t Hs

•MHASignal::waveform_t xw

•MHASignal::spectrum_t Xs

•MHASignal::waveform_t yw

•MHASignal::spectrum_t Ys

•MHASignal::waveform_t yw_temp

•MHASignal::waveform_t tail

•mha_fft_t fft

5.191.1 Detailed Description

FFT based FIR ﬁlterbank implementation.

This class convolves n input channels with m ﬁlter coefﬁcient sets and returns n∗m output

channels.

The maximal number of coefﬁcients can be FFT length - fragsize + 1.

5.191.2 Constructor & Destructor Documentation

5.191.2.1 MHAFilter::fftﬁlterbank_t::fftﬁlterbank_t (

unsigned int fragsize,

unsigned int inputchannels,

unsigned int ﬁrchannels,

unsigned int fftlen )

Constructor.

5.191 MHAFilter::fftﬁlterbank_t Class Reference 529

Parameters

fragsize Number of frames expected in input signal (each cycle).

inputchannels Number of channels expected in input signal.

ﬁrchannels Number of channels expected in FIR ﬁlter coefﬁcients (= number of bands).

fftlen FFT length of ﬁlter.

The number of output channels is inputchannels∗ﬁrchannels.

5.191.2.2 MHAFilter::fftﬁlterbank_t::∼fftﬁlterbank_t ( )

5.191.3 Member Function Documentation

5.191.3.1 void MHAFilter::fftﬁlterbank_t::update_coeffs (

const mha_wave_t ∗h)

Update the set of coefﬁcients.

Parameters

hCoefﬁcients structure

Note

The number of channels in h must match the number of channels given in the constructor,

and the number of frames can not be more than fftlen-fragsize+1.

5.191.3.2 void MHAFilter::fftﬁlterbank_t::ﬁlter (

const mha_wave_t ∗s_in,

mha_wave_t ∗∗ s_out,

const mha_wave_t ∗h)

Apply ﬁlter with changing coefﬁcients to a waveform fragment.

Parameters

s←-

_in

Input signal pointer.

Return values

s_out Pointer to output signal pointer, will be set to a valid signal

530 CONTENTS

Parameters

hFIR coefﬁcients

5.191.3.3 void MHAFilter::fftﬁlterbank_t::ﬁlter (

const mha_wave_t ∗s_in,

mha_wave_t ∗∗ s_out )

Apply ﬁlter to waveform fragment, without changing the coefﬁcients.

Parameters

s←-

_in

Input signal pointer.

Return values

s_out Pointer to output signal pointer, will be set to a valid signal

5.191.3.4 const mha_wave_t∗MHAFilter::fftﬁlterbank_t::get_irs ( ) const [inline]

Return the current IRS.

5.191.4 Member Data Documentation

5.191.4.1 unsigned int MHAFilter::fftﬁlterbank_t::fragsize [private]

5.191.4.2 unsigned int MHAFilter::fftﬁlterbank_t::inputchannels [private]

5.191.4.3 unsigned int MHAFilter::fftﬁlterbank_t::ﬁrchannels [private]

5.191.4.4 unsigned int MHAFilter::fftﬁlterbank_t::outputchannels [private]

5.191.4.5 unsigned int MHAFilter::fftﬁlterbank_t::fftlen [private]

5.191.4.6 MHASignal::waveform_t MHAFilter::fftﬁlterbank_t::hw [private]

5.191.4.7 MHASignal::spectrum_t MHAFilter::fftﬁlterbank_t::Hs [private]

5.191.4.8 MHASignal::waveform_t MHAFilter::fftﬁlterbank_t::xw [private]

5.191.4.9 MHASignal::spectrum_t MHAFilter::fftﬁlterbank_t::Xs [private]

5.192 MHAFilter::ﬁlter_t Class Reference 531

5.191.4.10 MHASignal::waveform_t MHAFilter::fftﬁlterbank_t::yw [private]

5.191.4.11 MHASignal::spectrum_t MHAFilter::fftﬁlterbank_t::Ys [private]

5.191.4.12 MHASignal::waveform_t MHAFilter::fftﬁlterbank_t::yw_temp [private]

5.191.4.13 MHASignal::waveform_t MHAFilter::fftﬁlterbank_t::tail [private]

5.191.4.14 mha_fft_t MHAFilter::fftﬁlterbank_t::fft [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.192 MHAFilter::ﬁlter_t Class Reference

Generic IIR ﬁlter class.

Inheritance diagram for MHAFilter::ﬁlter_t:

MHAFilter::filter_t

MHAFilter::diff_t MHAFilter::iir_filter

_state_t

Public Member Functions

•ﬁlter_t (unsigned int ch, unsigned int lena, unsigned int lenb)

Constructor.

•ﬁlter_t (unsigned int ch, const std::vector<mha_real_t >&vA, const std::vector<mha←-

_real_t >&vB)

Constructor with initialization of coefﬁcients.

•ﬁlter_t (const MHAFilter::ﬁlter_t &src)

•∼ﬁlter_t ()

• void ﬁlter (mha_wave_t ∗out, const mha_wave_t ∗in)

532 CONTENTS

Filter all channels in a waveform structure.

• void ﬁlter (mha_real_t ∗dest, const mha_real_t ∗src, unsigned int dframes, unsigned

int frame_dist, unsigned int channel_dist, unsigned int channel_begin, unsigned int

channel_end)

Filter parts of a waveform structure.

•mha_real_t ﬁlter (mha_real_t x, unsigned int ch)

Filter one sample.

• unsigned int get_len_A () const

Return length of recursive coefﬁcients.

• unsigned int get_len_B () const

Return length of non-recursive coefﬁcients.

Public Attributes

• double ∗A

Pointer to recursive coefﬁcients.

• double ∗B

Pointer to non-recursive coefﬁcients.

Private Attributes

• unsigned int len_A

• unsigned int len_B

• unsigned int len

• unsigned int channels

• double ∗state

5.192.1 Detailed Description

Generic IIR ﬁlter class.

This class implements a generic multichannel IIR ﬁlter. It is realized as direct form II. It can

work on any ﬂoat array or on mha_wave_t (p. 496) structs. The ﬁlter coefﬁcients can be directly

accessed.

Todo Implement a more robust ﬁlter form.

5.192.2 Constructor & Destructor Documentation

5.192.2.1 MHAFilter::ﬁlter_t::ﬁlter_t (

unsigned int ch,

unsigned int lena,

unsigned int lenb )

Constructor.

5.192 MHAFilter::ﬁlter_t Class Reference 533

Parameters

ch Number of channels

lena Number of recursive coefﬁcients

lenb Number of non-recursive coefﬁcients

5.192.2.2 MHAFilter::ﬁlter_t::ﬁlter_t (

unsigned int ch,

const std::vector<mha_real_t >&vA,

const std::vector<mha_real_t >&vB )

Constructor with initialization of coefﬁcients.

Parameters

ch Number of channels.

vA Recursive coefﬁcients.

vB Non-recursive coefﬁcients.

5.192.2.3 MHAFilter::ﬁlter_t::ﬁlter_t (

const MHAFilter::ﬁlter_t &src )

5.192.2.4 MHAFilter::ﬁlter_t::∼ﬁlter_t ( )

5.192.3 Member Function Documentation

5.192.3.1 void MHAFilter::ﬁlter_t::ﬁlter (

mha_wave_t ∗out,

const mha_wave_t ∗in )

Filter all channels in a waveform structure.

Parameters

out Output signal

in Input signal

534 CONTENTS

5.192.3.2 void MHAFilter::ﬁlter_t::ﬁlter (

mha_real_t ∗dest,

const mha_real_t ∗src,

unsigned int dframes,

unsigned int frame_dist,

unsigned int channel_dist,

unsigned int channel_begin,

unsigned int channel_end )

Filter parts of a waveform structure.

Parameters

dest Output signal.

src Input signal.

dframes Number of frames to be ﬁltered.

frame_dist Index distance between frames of one channel

channel_dist Index distance between audio channels

channel_begin Number of ﬁrst channel to be processed

channel_end Number of last channel to be processed

5.192.3.3 mha_real_t MHAFilter::ﬁlter_t::ﬁlter (

mha_real_t x,

unsigned int ch )

Filter one sample.

Parameters

xInput value

ch Channel number to use in ﬁlter state

5.192.3.4 unsigned int MHAFilter::ﬁlter_t::get_len_A ( ) const [inline]

Return length of recursive coefﬁcients.

5.192.3.5 unsigned int MHAFilter::ﬁlter_t::get_len_B ( ) const [inline]

Return length of non-recursive coefﬁcients.

5.192.4 Member Data Documentation

5.192.4.1 double∗MHAFilter::ﬁlter_t::A

Pointer to recursive coefﬁcients.

5.193 MHAFilter::gamma_ﬂt_t Class Reference 535

5.192.4.2 double∗MHAFilter::ﬁlter_t::B

Pointer to non-recursive coefﬁcients.

5.192.4.3 unsigned int MHAFilter::ﬁlter_t::len_A [private]

5.192.4.4 unsigned int MHAFilter::ﬁlter_t::len_B [private]

5.192.4.5 unsigned int MHAFilter::ﬁlter_t::len [private]

5.192.4.6 unsigned int MHAFilter::ﬁlter_t::channels [private]

5.192.4.7 double∗MHAFilter::ﬁlter_t::state [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.193 MHAFilter::gamma_ﬂt_t Class Reference

Class for gammatone ﬁlter.

Public Member Functions

•gamma_ﬂt_t (std::vector<mha_real_t >cf, std::vector<mha_real_t >bw, mha_real←-

_t srate, unsigned int order)

Constructor.

•∼gamma_ﬂt_t ()

• void operator() (mha_wave_t &X, mha_spec_t &Y)

Filter method.

• void operator() (mha_wave_t &X, mha_wave_t &Yre, mha_wave_t &Yim)

Filter method.

• void operator() (mha_wave_t &Yre, mha_wave_t &Yim, unsigned int stage)

Filter method for speciﬁc stage.

• void phase_correction (unsigned int desired_delay, unsigned int inchannels)

• void set_weights (std::vector<mha_complex_t >new_B)

• void set_weights (unsigned int stage, std::vector<mha_complex_t >new_B)

• std::vector<mha_complex_t >get_weights () const

• std::vector<mha_complex_t >get_weights (unsigned int stage) const

• std::vector<mha_real_t >get_resynthesis_gain () const

• void reset_state ()

• const std::vector<mha_complex_t >&get_A ()

• std::string inspect () const

536 CONTENTS

Private Attributes

• std::vector<mha_complex_t >A

• std::vector<complex_bandpass_t >GF

•MHASignal::delay_t ∗delay

• std::vector<int >envelope_delay

• std::vector<mha_real_t >resynthesis_gain

• std::vector<mha_real_t >cf_

• std::vector<mha_real_t >bw_

•mha_real_t srate_

5.193.1 Detailed Description

Class for gammatone ﬁlter.

5.193.2 Constructor & Destructor Documentation

5.193.2.1 MHAFilter::gamma_ﬂt_t::gamma_ﬂt_t (

std::vector<mha_real_t >cf,

std::vector<mha_real_t >bw,

mha_real_t srate,

unsigned int order )

Constructor.

Parameters

cf Center frequency in Hz.

bw Bandwidth in Hz (same number of entries as in cf).

srate Sampling frequency in Hz.

order Filter order.

5.193.2.2 MHAFilter::gamma_ﬂt_t::∼gamma_ﬂt_t ( )

5.193.3 Member Function Documentation

5.193.3.1 void MHAFilter::gamma_ﬂt_t::operator() (

mha_wave_t &X,

mha_spec_t &Y)[inline]

Filter method.

5.193 MHAFilter::gamma_ﬂt_t Class Reference 537

5.193.3.2 void MHAFilter::gamma_ﬂt_t::operator() (

mha_wave_t &X,

mha_wave_t &Yre,

mha_wave_t &Yim )[inline]

Filter method.

5.193.3.3 void MHAFilter::gamma_ﬂt_t::operator() (

mha_wave_t &Yre,

mha_wave_t &Yim,

unsigned int stage )[inline]

Filter method for speciﬁc stage.

5.193.3.4 void MHAFilter::gamma_ﬂt_t::phase_correction (

unsigned int desired_delay,

unsigned int inchannels )

5.193.3.5 void MHAFilter::gamma_ﬂt_t::set_weights (

std::vector<mha_complex_t >new_B )

5.193.3.6 void MHAFilter::gamma_ﬂt_t::set_weights (

unsigned int stage,

std::vector<mha_complex_t >new_B )

5.193.3.7 std::vector<mha_complex_t>MHAFilter::gamma_ﬂt_t::get_weights ( ) const

[inline]

5.193.3.8 std::vector<mha_complex_t>MHAFilter::gamma_ﬂt_t::get_weights (

unsigned int stage ) const [inline]

5.193.3.9 std::vector<mha_real_t>MHAFilter::gamma_ﬂt_t::get_resynthesis_gain ( ) const

[inline]

5.193.3.10 void MHAFilter::gamma_ﬂt_t::reset_state ( )

5.193.3.11 const std::vector<mha_complex_t>& MHAFilter::gamma_ﬂt_t::get_A ( ) [inline]

5.193.3.12 std::string MHAFilter::gamma_ﬂt_t::inspect ( ) const [inline]

5.193.4 Member Data Documentation

5.193.4.1 std::vector<mha_complex_t>MHAFilter::gamma_ﬂt_t::A [private]

5.193.4.2 std::vector<complex_bandpass_t>MHAFilter::gamma_ﬂt_t::GF [private]

538 CONTENTS

5.193.4.3 MHASignal::delay_t∗MHAFilter::gamma_ﬂt_t::delay [private]

5.193.4.4 std::vector<int>MHAFilter::gamma_ﬂt_t::envelope_delay [private]

5.193.4.5 std::vector<mha_real_t>MHAFilter::gamma_ﬂt_t::resynthesis_gain [private]

5.193.4.6 std::vector<mha_real_t>MHAFilter::gamma_ﬂt_t::cf_ [private]

5.193.4.7 std::vector<mha_real_t>MHAFilter::gamma_ﬂt_t::bw_ [private]

5.193.4.8 mha_real_t MHAFilter::gamma_ﬂt_t::srate_ [private]

The documentation for this class was generated from the following ﬁles:

•complex_ﬁlter.h

•complex_ﬁlter.cpp

5.194 MHAFilter::iir_ﬁlter_state_t Class Reference

Inheritance diagram for MHAFilter::iir_ﬁlter_state_t:

MHAFilter::iir_filter

_state_t

MHAFilter::filter_t

Public Member Functions

•iir_ﬁlter_state_t (unsigned int channels, std::vector<ﬂoat >cf_A, std::vector<ﬂoat >

cf_B)

5.195 MHAFilter::iir_ﬁlter_t Class Reference 539

Additional Inherited Members

5.194.1 Constructor & Destructor Documentation

5.194.1.1 MHAFilter::iir_ﬁlter_state_t::iir_ﬁlter_state_t (

unsigned int channels,

std::vector<ﬂoat >cf_A,

std::vector<ﬂoat >cf_B )

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.195 MHAFilter::iir_ﬁlter_t Class Reference

IIR ﬁlter class wrapper for integration into parser structure.

Inheritance diagram for MHAFilter::iir_ﬁlter_t:

MHAFilter::iir_filter_t

iirfilter_t

MHAParser::parser_t

MHAParser::base_t

MHAPlugin::config_t

< iir_filter_state_t >

540 CONTENTS

Public Member Functions

•iir_ﬁlter_t (std::string help="IIR ﬁlter structure", std::string def_A="[1]", std::string def←-

_B="[1]", unsigned int channels=1)

Constructor of the IIR ﬁlter.

• void ﬁlter (mha_wave_t ∗y, const mha_wave_t ∗x)

The ﬁlter processes the audio signal.

•mha_real_t ﬁlter (mha_real_t x, unsigned int ch)

Filter a single audio sample.

• void resize (unsigned int channels)

Change the number of channels after object creation.

Private Member Functions

• void update_ﬁlter ()

Private Attributes

•MHAParser::vﬂoat_t A

•MHAParser::vﬂoat_t B

•MHAEvents::patchbay_t<iir_ﬁlter_t >connector

• unsigned int nchannels

Additional Inherited Members

5.195.1 Detailed Description

IIR ﬁlter class wrapper for integration into parser structure.

This class implements an inﬁnite impulse response ﬁlter. Since it inherits from MHAParser←-

::parser_t (p. 688), it can easily be integrated in the openMHA conﬁguration tree. It provides

the conﬁguration language variables "A" (vector of recursive ﬁlter coefﬁcients) and "B" (vector

of non-recursive ﬁlter coefﬁcients).

The ﬁlter instance reacts to changes in ﬁlter coefﬁcients through the openMHA conﬁguration

language, and uses the updated coefﬁcients in the next invocation of the ﬁlter method.

Update of the coefﬁcients is thread-safe and non-blocking. Simply add this subparser to your

parser items and use the "ﬁlter" member function. Filter states are reset to all 0 on update.

5.195.2 Constructor & Destructor Documentation

5.195.2.1 MHAFilter::iir_ﬁlter_t::iir_ﬁlter_t (

std::string help = "IIR ﬁlter structure",

std::string def_A = "[1]",

std::string def_B = "[1]",

unsigned int channels = 1)

Constructor of the IIR ﬁlter.

Initialises the sub-parser structure and the memory for holding the ﬁlter's state.

5.195 MHAFilter::iir_ﬁlter_t Class Reference 541

Parameters

help The help string for the parser that groups the conﬁguration variables of this ﬁlter.

Could be used to describe the purpose of this IIR ﬁlter.

def_A The initial value of the vector of the recursive ﬁlter coefﬁcients, represented as

string.

def_B The initial value of the vector of the non-recursive ﬁlter coefﬁcients, represented

as string.

channels The number of indipendent audio channels to process with this ﬁlter. Needed to

allocate a state vector for each audio channel.

5.195.3 Member Function Documentation

5.195.3.1 void MHAFilter::iir_ﬁlter_t::ﬁlter (

mha_wave_t ∗y,

const mha_wave_t ∗x)

The ﬁlter processes the audio signal.

All channels in the audio signal are processed using the same ﬁlter coefﬁcients. Indipendent

state is stored between calls for each audio channel.

Parameters

yPointer to output signal holder. The output signal is stored here. Has to have the same

signal dimensions as the input signal x. In-place processing (y and x pointing to the

same signal holder) is possible.

xPointer to input signal holder. Number of channels has to be the same as given to the

constructor, or to the resize (p. 542) method.

5.195.3.2 mha_real_t MHAFilter::iir_ﬁlter_t::ﬁlter (

mha_real_t x,

unsigned int ch )

Filter a single audio sample.

Parameters

xThe single audio sample

ch Zero-based channel index. Use and change the state of channel ch. ch has to be less

than the number of channels given to the constructor or the resize (p. 542) method.

542 CONTENTS

Returns

the ﬁltered result sample.

5.195.3.3 void MHAFilter::iir_ﬁlter_t::resize (

unsigned int channels )

Change the number of channels after object creation.

Parameters

channels The new number of channels. Old ﬁlter states are lost.

5.195.3.4 void MHAFilter::iir_ﬁlter_t::update_ﬁlter ( ) [private]

5.195.4 Member Data Documentation

5.195.4.1 MHAParser::vﬂoat_t MHAFilter::iir_ﬁlter_t::A [private]

5.195.4.2 MHAParser::vﬂoat_t MHAFilter::iir_ﬁlter_t::B [private]

5.195.4.3 MHAEvents::patchbay_t<iir_ﬁlter_t>MHAFilter::iir_ﬁlter_t::connector [private]

5.195.4.4 unsigned int MHAFilter::iir_ﬁlter_t::nchannels [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.196 MHAFilter::iir_ord1_real_t Class Reference

First order recursive ﬁlter.

5.196 MHAFilter::iir_ord1_real_t Class Reference 543

Public Member Functions

•iir_ord1_real_t (std::vector<mha_real_t >A, std::vector<mha_real_t >B)

Constructor with ﬁlter coefﬁcients (one per channel)

•iir_ord1_real_t (std::vector<mha_real_t >tau, mha_real_t srate)

Constructor for low pass ﬁlter (one time constant per channel)

• void set_state (mha_real_t val)

• void set_state (std::vector<mha_real_t >val)

• void set_state (mha_complex_t val)

•mha_real_t operator() (unsigned int ch, mha_real_t x)

Filter method for real value input, one element.

•mha_complex_t operator() (unsigned int ch, mha_complex_t x)

Filter method for complex input, one element.

• void operator() (const mha_wave_t &X, mha_wave_t &Y)

Filter method for real value input.

• void operator() (const mha_spec_t &X, mha_spec_t &Y)

Filter method for complex value input.

• void operator() (const mha_wave_t &Xre, const mha_wave_t &Xim, mha_wave_t &Yre,

mha_wave_t &Yim)

Filter method for complex value input.

Private Attributes

• std::vector<mha_real_t >A_

• std::vector<mha_real_t >B_

• std::vector<mha_complex_t >Yn

5.196.1 Detailed Description

First order recursive ﬁlter.

5.196.2 Constructor & Destructor Documentation

5.196.2.1 MHAFilter::iir_ord1_real_t::iir_ord1_real_t (

std::vector<mha_real_t >A,

std::vector<mha_real_t >B)

Constructor with ﬁlter coefﬁcients (one per channel)

5.196.2.2 MHAFilter::iir_ord1_real_t::iir_ord1_real_t (

std::vector<mha_real_t >tau,

mha_real_t srate )

Constructor for low pass ﬁlter (one time constant per channel)

544 CONTENTS

5.196.3 Member Function Documentation

5.196.3.1 void MHAFilter::iir_ord1_real_t::set_state (

mha_real_t val )

5.196.3.2 void MHAFilter::iir_ord1_real_t::set_state (

std::vector<mha_real_t >val )

5.196.3.3 void MHAFilter::iir_ord1_real_t::set_state (

mha_complex_t val )

5.196.3.4 mha_real_t MHAFilter::iir_ord1_real_t::operator() (

unsigned int ch,

mha_real_t x)[inline]

Filter method for real value input, one element.

5.196.3.5 mha_complex_t MHAFilter::iir_ord1_real_t::operator() (

unsigned int ch,

mha_complex_t x)[inline]

Filter method for complex input, one element.

5.196.3.6 void MHAFilter::iir_ord1_real_t::operator() (

const mha_wave_t &X,

mha_wave_t &Y)[inline]

Filter method for real value input.

5.196.3.7 void MHAFilter::iir_ord1_real_t::operator() (

const mha_spec_t &X,

mha_spec_t &Y)[inline]

Filter method for complex value input.

5.196.3.8 void MHAFilter::iir_ord1_real_t::operator() (

const mha_wave_t &Xre,

const mha_wave_t &Xim,

mha_wave_t &Yre,

mha_wave_t &Yim )[inline]

Filter method for complex value input.

5.197 MHAFilter::o1_ar_ﬁlter_t Class Reference 545

5.196.4 Member Data Documentation

5.196.4.1 std::vector<mha_real_t>MHAFilter::iir_ord1_real_t::A_ [private]

5.196.4.2 std::vector<mha_real_t>MHAFilter::iir_ord1_real_t::B_ [private]

5.196.4.3 std::vector<mha_complex_t>MHAFilter::iir_ord1_real_t::Yn [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.197 MHAFilter::o1_ar_ﬁlter_t Class Reference

First order attack-release lowpass ﬁlter.

Inheritance diagram for MHAFilter::o1_ar_ﬁlter_t:

MHAFilter::o1_ar_filter_t

MHAFilter::o1flt_lowpass_t

MHASignal::waveform_t

mha_wave_t

MHAFilter::o1flt_maxtrack_t MHAFilter::o1flt_mintrack_t

546 CONTENTS

Public Member Functions

•o1_ar_ﬁlter_t (unsigned int channels,mha_real_t fs=1.0f, std::vector<mha_real_t >

tau_a=std::vector<ﬂoat >(1, 0.0f), std::vector<mha_real_t >tau_r=std::vector<ﬂoat

>(1, 0.0f))

Constructor, setting all taus to zero.

• void set_tau_attack (unsigned int ch, mha_real_t tau)

Set the attack time constant.

• void set_tau_release (unsigned int ch, mha_real_t tau)

Set the release time constant.

•mha_real_t operator() (unsigned int ch, mha_real_t x)

Apply ﬁlter to value x, using state channel ch.

• void operator() (const mha_wave_t &in, mha_wave_t &out)

Apply ﬁlter to a mha_wave_t (p. 496) data.

Protected Attributes

•MHASignal::waveform_t c1_a

•MHASignal::waveform_t c2_a

•MHASignal::waveform_t c1_r

•MHASignal::waveform_t c2_r

•mha_real_t fs

Additional Inherited Members

5.197.1 Detailed Description

First order attack-release lowpass ﬁlter.

This ﬁlter is the base of ﬁrst order lowpass ﬁlter, maximum tracker and minimum tracker.

5.197.2 Constructor & Destructor Documentation

5.197.2.1 MHAFilter::o1_ar_ﬁlter_t::o1_ar_ﬁlter_t (

unsigned int channels,

mha_real_t fs = 1.0f,

std::vector<mha_real_t >tau_a = std::vector<float>(1,0.0f),

std::vector<mha_real_t >tau_r = std::vector<float>(1,0.0f) )

Constructor, setting all taus to zero.

The ﬁlter state can be accessed through the member functions of MHASignal::waveform_t

(p. 812).

5.197 MHAFilter::o1_ar_ﬁlter_t Class Reference 547

Parameters

channels Number of independent ﬁlters

fs Sampling rate (optional, default = 1)

tau_a Attack time constants (optional, default = 0)

tau_r Release time constants (optional, default = 0)

5.197.3 Member Function Documentation

5.197.3.1 void MHAFilter::o1_ar_ﬁlter_t::set_tau_attack (

unsigned int ch,

mha_real_t tau )

Set the attack time constant.

Parameters

ch Channel number

tau Time constant

5.197.3.2 void MHAFilter::o1_ar_ﬁlter_t::set_tau_release (

unsigned int ch,

mha_real_t tau )

Set the release time constant.

Parameters

ch Channel number

tau Time constant

5.197.3.3 mha_real_t MHAFilter::o1_ar_ﬁlter_t::operator() (

unsigned int ch,

mha_real_t x)[inline]

Apply ﬁlter to value x, using state channel ch.

Parameters

ch Cannel number

xInput value

548 CONTENTS

Returns

Output value

5.197.3.4 void MHAFilter::o1_ar_ﬁlter_t::operator() (

const mha_wave_t &in,

mha_wave_t &out )[inline]

Apply ﬁlter to a mha_wave_t (p. 496) data.

Parameters

in Input signal

out Output signal

The number of channels must match the number of ﬁlter bands.

5.197.4 Member Data Documentation

5.197.4.1 MHASignal::waveform_t MHAFilter::o1_ar_ﬁlter_t::c1_a [protected]

5.197.4.2 MHASignal::waveform_t MHAFilter::o1_ar_ﬁlter_t::c2_a [protected]

5.197.4.3 MHASignal::waveform_t MHAFilter::o1_ar_ﬁlter_t::c1_r [protected]

5.197.4.4 MHASignal::waveform_t MHAFilter::o1_ar_ﬁlter_t::c2_r [protected]

5.197.4.5 mha_real_t MHAFilter::o1_ar_ﬁlter_t::fs [protected]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.198 MHAFilter::o1ﬂt_lowpass_t Class Reference

First order low pass ﬁlter.

5.198 MHAFilter::o1ﬂt_lowpass_t Class Reference 549

Inheritance diagram for MHAFilter::o1ﬂt_lowpass_t:

MHAFilter::o1flt_lowpass_t

MHAFilter::o1flt_maxtrack_t MHAFilter::o1flt_mintrack_t

MHAFilter::o1_ar_filter_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•o1ﬂt_lowpass_t (const std::vector<mha_real_t >&, mha_real_t,mha_real_t=0)

Constructor of low pass ﬁlter, sets sampling rate and time constants.

• void set_tau (unsigned int ch, mha_real_t tau)

change the time constant in one channel

• void set_tau (mha_real_t tau)

set time constant in all channels to tau

•mha_real_t get_c1 (unsigned int ch) const

•mha_real_t get_last_output (unsigned int ch) const

Additional Inherited Members

5.198.1 Detailed Description

First order low pass ﬁlter.

550 CONTENTS

5.198.2 Constructor & Destructor Documentation

5.198.2.1 MHAFilter::o1ﬂt_lowpass_t::o1ﬂt_lowpass_t (

const std::vector<mha_real_t >&tau,

mha_real_t fs,

mha_real_t startval = 0)

Constructor of low pass ﬁlter, sets sampling rate and time constants.

Parameters

tau Vector of time constants

fs Sampling rate

startval Initial internal state value

5.198.3 Member Function Documentation

5.198.3.1 void MHAFilter::o1ﬂt_lowpass_t::set_tau (

unsigned int ch,

mha_real_t tau )

change the time constant in one channel

5.198.3.2 void MHAFilter::o1ﬂt_lowpass_t::set_tau (

mha_real_t tau )

set time constant in all channels to tau

5.198.3.3 mha_real_t MHAFilter::o1ﬂt_lowpass_t::get_c1 (

unsigned int ch ) const [inline]

5.198.3.4 mha_real_t MHAFilter::o1ﬂt_lowpass_t::get_last_output (

unsigned int ch ) const [inline]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.199 MHAFilter::o1ﬂt_maxtrack_t Class Reference 551

5.199 MHAFilter::o1ﬂt_maxtrack_t Class Reference

First order maximum tracker.

Inheritance diagram for MHAFilter::o1ﬂt_maxtrack_t:

MHAFilter::o1flt_maxtrack_t

MHAFilter::o1flt_lowpass_t

MHAFilter::o1_ar_filter_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•o1ﬂt_maxtrack_t (const std::vector<mha_real_t >&, mha_real_t,mha_real_t=0)

Constructor of low pass ﬁlter, sets sampling rate and time constants.

• void set_tau (unsigned int ch, mha_real_t tau)

change the time constant in one channel

• void set_tau (mha_real_t tau)

set time constant in all channels to tau

Additional Inherited Members

5.199.1 Detailed Description

First order maximum tracker.

552 CONTENTS

5.199.2 Constructor & Destructor Documentation

5.199.2.1 MHAFilter::o1ﬂt_maxtrack_t::o1ﬂt_maxtrack_t (

const std::vector<mha_real_t >&tau,

mha_real_t fs,

mha_real_t startval = 0)

Constructor of low pass ﬁlter, sets sampling rate and time constants.

Parameters

tau Vector of time constants

fs Sampling rate

startval Initial internal state value

5.199.3 Member Function Documentation

5.199.3.1 void MHAFilter::o1ﬂt_maxtrack_t::set_tau (

unsigned int ch,

mha_real_t tau )

change the time constant in one channel

5.199.3.2 void MHAFilter::o1ﬂt_maxtrack_t::set_tau (

mha_real_t tau )

set time constant in all channels to tau

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.200 MHAFilter::o1ﬂt_mintrack_t Class Reference

First order minimum tracker.

5.200 MHAFilter::o1ﬂt_mintrack_t Class Reference 553

Inheritance diagram for MHAFilter::o1ﬂt_mintrack_t:

MHAFilter::o1flt_mintrack_t

MHAFilter::o1flt_lowpass_t

MHAFilter::o1_ar_filter_t

MHASignal::waveform_t

mha_wave_t

Public Member Functions

•o1ﬂt_mintrack_t (const std::vector<mha_real_t >&, mha_real_t,mha_real_t=0)

• void set_tau (unsigned int ch, mha_real_t tau)

change the time constant in one channel

• void set_tau (mha_real_t tau)

set time constant in all channels to tau

Additional Inherited Members

5.200.1 Detailed Description

First order minimum tracker.

554 CONTENTS

5.200.2 Constructor & Destructor Documentation

5.200.2.1 MHAFilter::o1ﬂt_mintrack_t::o1ﬂt_mintrack_t (

const std::vector<mha_real_t >&tau,

mha_real_t fs_,

mha_real_t startval = 0)

5.200.3 Member Function Documentation

5.200.3.1 void MHAFilter::o1ﬂt_mintrack_t::set_tau (

unsigned int ch,

mha_real_t tau )

change the time constant in one channel

5.200.3.2 void MHAFilter::o1ﬂt_mintrack_t::set_tau (

mha_real_t tau )

set time constant in all channels to tau

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.201 MHAFilter::partitioned_convolution_t Class Reference

A ﬁlter class for partitioned convolution.

Classes

• struct index_t

Bookkeeping class.

Public Member Functions

•partitioned_convolution_t (unsigned int fragsize, unsigned int nchannels_in, unsigned

int nchannels_out, const transfer_matrix_t &transfer)

Create a new partitioned convolver.

•∼partitioned_convolution_t ()

Free fftw resource allocated in constructor.

•mha_wave_t ∗process (const mha_wave_t ∗s_in)

processing

5.201 MHAFilter::partitioned_convolution_t Class Reference 555

Public Attributes

• unsigned int fragsize

Audio fragment size, always equal to partition size.

• unsigned int nchannels_in

Number of audio input channels.

• unsigned int nchannels_out

Number of audio output channels.

• unsigned int output_partitions

The maximum number of partitions in any of the impulse responses.

• unsigned int ﬁlter_partitions

The total number of non-zero impulse response partitions.

•MHASignal::waveform_t input_signal_wave

Buffer for input signal.

• unsigned int current_input_signal_buffer_half_index

A counter modulo 2.

•MHASignal::spectrum_t input_signal_spec

Buffer for FFT transformed input signal.

•MHASignal::spectrum_t frequency_response

Buffers for frequency response spectra of impulse response partitions.

• std::vector<index_t >bookkeeping

Keeps track of input channels, output channels, impulse response partition, and delay.

• std::vector<MHASignal::spectrum_t >output_signal_spec

Buffers for FFT transformed output signal.

• unsigned int current_output_partition_index

A counter modulo output_partitions, indexing the "current" output partition.

•MHASignal::waveform_t output_signal_wave

Buffer for the wave output signal.

•mha_fft_t fft

The FFT transformer.

5.201.1 Detailed Description

A ﬁlter class for partitioned convolution.

Impulse responses are partitioned into sections of fragment size. Audio signal is convolved with

every partition and delayed as needed. Convolution is done according to overlap-save. FFT

length used is 2 times fragment size.

5.201.2 Constructor & Destructor Documentation

5.201.2.1 MHAFilter::partitioned_convolution_t::partitioned_convolution_t (

unsigned int fragsize,

unsigned int nchannels_in,

unsigned int nchannels_out,

const transfer_matrix_t &transfer )

Create a new partitioned convolver.

556 CONTENTS

Parameters

fragsize Audio fragment size, equal to partition size.

nchannels_in Number of input audio channels.

nchannels_out Number of output audio channels.

transfer A sparse matrix of impulse responses.

5.201.2.2 MHAFilter::partitioned_convolution_t::∼partitioned_convolution_t ( )

Free fftw resource allocated in constructor.

5.201.3 Member Function Documentation

5.201.3.1 mha_wave_t ∗MHAFilter::partitioned_convolution_t::process (

const mha_wave_t ∗s_in )

processing

5.201.4 Member Data Documentation

5.201.4.1 unsigned int MHAFilter::partitioned_convolution_t::fragsize

Audio fragment size, always equal to partition size.

5.201.4.2 unsigned int MHAFilter::partitioned_convolution_t::nchannels_in

Number of audio input channels.

5.201.4.3 unsigned int MHAFilter::partitioned_convolution_t::nchannels_out

Number of audio output channels.

5.201.4.4 unsigned int MHAFilter::partitioned_convolution_t::output_partitions

The maximum number of partitions in any of the impulse responses.

Determines the size if the delay line.

5.201.4.5 unsigned int MHAFilter::partitioned_convolution_t::ﬁlter_partitions

The total number of non-zero impulse response partitions.

5.201 MHAFilter::partitioned_convolution_t Class Reference 557

5.201.4.6 MHASignal::waveform_t MHAFilter::partitioned_convolution_t::input_signal_wave

Buffer for input signal.

Has nchannels_in channels and fragsize∗2 frames

5.201.4.7 unsigned int MHAFilter::partitioned_convolution_t::current_input_signal_buffer_half_index

A counter modulo 2.

Indicates the buffer half in input signal wave into which to copy the current input signal.

5.201.4.8 MHASignal::spectrum_t MHAFilter::partitioned_convolution_t::input_signal_spec

Buffer for FFT transformed input signal.

Has nchannels_in channels and fragsize+1 frames (fft bins).

5.201.4.9 MHASignal::spectrum_t MHAFilter::partitioned_convolution_t::frequency_response

Buffers for frequency response spectra of impulse response partitions.

Each "channel" contains another partition of some impulse response. The bookkeeping array

is used to keep track what to do with these frequency responses. This container has ﬁlter_←-

partitions channels and fragsize+1 frames (fft bins).

5.201.4.10 std::vector<index_t>MHAFilter::partitioned_convolution_t::bookkeeping

Keeps track of input channels, output channels, impulse response partition, and delay.

The index into this array is the same as the "channel" index into the frequency_response array.

Array has ﬁlter_partitions entries.

5.201.4.11 std::vector<MHASignal::spectrum_t>MHAFilter::partitioned_convolution_t::output_←-

signal_spec

Buffers for FFT transformed output signal.

For each array member, Number of channels is equal to nchannels_out, number of frames (fft

bins) is equal to fragsize+1. Array size is equal to output_partitions.

5.201.4.12 unsigned int MHAFilter::partitioned_convolution_t::current_output_partition_index

A counter modulo output_partitions, indexing the "current" output partition.

558 CONTENTS

5.201.4.13 MHASignal::waveform_t MHAFilter::partitioned_convolution_t::output_signal_wave

Buffer for the wave output signal.

Number of channels is equal to nchannels_out, number of frames is equal to fragsize

5.201.4.14 mha_fft_t MHAFilter::partitioned_convolution_t::fft

The FFT transformer.

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.202 MHAFilter::partitioned_convolution_t::index_t Struct Reference

Bookkeeping class.

Public Member Functions

•index_t (unsigned int src, unsigned int tgt, unsigned int dly)

Data constructor.

•index_t ()

Default constructor for STL compatibility.

Public Attributes

• unsigned int source_channel_index

The input channel index to apply the current partition to.

• unsigned int target_channel_index

The index of the output channel to which the ﬁlter result should go.

• unsigned int delay

The delay (in blocks) of this partition.

5.202.1 Detailed Description

Bookkeeping class.

For each impulse response partition, keeps track of which input to ﬁlter, which output channel

to ﬁlter to, and the delay in blocks. Objects of class Index should be kept in an array with the

same indices as the corresponding inpulse response partitions.

5.202.2 Constructor & Destructor Documentation

5.202.2.1 MHAFilter::partitioned_convolution_t::index_t::index_t (

unsigned int src,

unsigned int tgt,

unsigned int dly )[inline]

Data constructor.

5.203 MHAFilter::polyphase_resampling_t Class Reference 559

Parameters

src The input channel index to apply the current partition to.

tgt The index of the output channel to which the ﬁlter result should go.

dly The delay (in blocks) of this partition

5.202.2.2 MHAFilter::partitioned_convolution_t::index_t::index_t ( ) [inline]

Default constructor for STL compatibility.

5.202.3 Member Data Documentation

5.202.3.1 unsigned int MHAFilter::partitioned_convolution_t::index_t::source_channel_index

The input channel index to apply the current partition to.

5.202.3.2 unsigned int MHAFilter::partitioned_convolution_t::index_t::target_channel_index

The index of the output channel to which the ﬁlter result should go.

5.202.3.3 unsigned int MHAFilter::partitioned_convolution_t::index_t::delay

The delay (in blocks) of this partition.

The documentation for this struct was generated from the following ﬁle:

•mha_ﬁlter.hh

5.203 MHAFilter::polyphase_resampling_t Class Reference

A class that performs polyphase resampling.

Public Member Functions

•polyphase_resampling_t (unsigned n_up, unsigned n_down, mha_real_t nyquist_ratio,

unsigned n_irs, unsigned n_ringbuffer, unsigned n_channels, unsigned n_preﬁll)

Construct a polyphase resampler instance.

• void write (mha_wave_t &signal)

Write signal to the ringbuffer.

• void read (mha_wave_t &signal)

Read resampled signal.

• unsigned readable_frames () const

Number of frames at target sampling rate that can be produced.

560 CONTENTS

Private Attributes

• unsigned upsampling_factor

Integer upsampling factor.

• unsigned downsampling_factor

Integer downsampling factor.

• unsigned now_index

Index of "now" in the interpolated sampling rate.

• bool underﬂow

Set to true when an underﬂow has occurred.

•MHAWindow::hanning_t impulse_response

Contains the impulse response of the lowpass ﬁlter needed for anti-aliasing.

•MHASignal::ringbuffer_t ringbuffer

Storage of input signal.

5.203.1 Detailed Description

A class that performs polyphase resampling.

Background information: When resampling from one sampling rate to another, it helps when

one sampling rate is a multiple of the other sampling rate: In the case of upsampling, the

samples at the original rate are copied to the upsampled signal spread out with a constant

number of zero samples between the originally adjacent samples. The signal is then low-pass

ﬁltered to avoid frequency aliasing and to ﬁll the zero-samples with interpolated values. In the

case of down-sampling, the signal is ﬁrst low-pass ﬁltered for anti-aliasing, and only every nth

sample of the ﬁltered output is used for the signal at the new sample rate. Of course, for ﬁnite-

impulse-response (FIR) ﬁlters this means that only every nth sample needs to be computed.

When resampling from one sampling rate to another where neither is a multiple of the other,

the signal ﬁrst needs to be upsampled to a sampling rate that is a multiple of both (source and

target) sampling rates, and then downsampled again to the target sampling rate. Instead of

applying two separate lowpass ﬁlters directly after each other (one ﬁlter for upsampling and

another for downsampling), it is sufﬁcient to apply only one low-pass ﬁlter, when producing

the output at the ﬁnal target rate, with a cut-off frequency equal to the lower cut-off-frequency

of the replaced two low-pass ﬁlters. Not ﬁltering to produce a ﬁltered signal already at the

common multiple sampling rate has the side effect that this intermediate signal at the common

multiple sampling rate keeps its ﬁller zero samples unaltered. These zero samples can be taken

advantage of when ﬁltering to produce the output at the target rate: The zeros do not need to

be multiplied with their corresponding ﬁlter coefﬁcients, because the result is known to be zero

again, and this zero product has no effect on the summation operation to compute a target

sample at the target rate. To summarize, the following optimization techniques are available:

• The signal does not need to be stored in memory at the interpolation rate. It is sufﬁcient

to have the signal available at the source rate and to know where the zeros would be.

• The signal needs to be low-pass-ﬁltered only once.

• The FIR low-pass ﬁltering can take advantage of

5.203 MHAFilter::polyphase_resampling_t Class Reference 561

–computing only ﬁlter outputs for the required samples at the target rate,

–skipping over zero-samples at the interpolation rate.

The procedure that takes advantage of these optimization possibilites is known as polyphase

resampling.

This class implements polyphase resampling in this way for a source sampling rate and a

target sampling rate that have common multiple, the interpolation sampling rate. Non-rational

and drifting sample rates are outside the scope of this resampler.

5.203.2 Constructor & Destructor Documentation

5.203.2.1 MHAFilter::polyphase_resampling_t::polyphase_resampling_t (

unsigned n_up,

unsigned n_down,

mha_real_t nyquist_ratio,

unsigned n_irs,

unsigned n_ringbuffer,

unsigned n_channels,

unsigned n_preﬁll )

Construct a polyphase resampler instance.

Allocates a ringbuffer with the given capacity n_ringbuffer. Client that triggers the constructor

must ensure that the capacity n_ringbuffer and the delay n_preﬁll are sufﬁcient, i.e. enough

old and new samples are always available to compute sufﬁcient samples in using an impulse

response of length n_irs. Audio block sizes at both sides of the resampler have to be taken into

account. Class MHASignal::blockprocessing_polyphase_resampling_t takes care

of this, and it is recommended to use this class for block-based processing.

Based on n_up,n_down,n_irs and nyquist_ratio, a suitable sinc impulse response is computed

and windowed with a hanning window to limit its extent.

The actual source sampling rate, target sampling rate, and interpolation sampling rate are not

parameters to this constructors, because only their ratios matter.

Parameters

n_up upsampling factor, ratio between interpolation rate and source rate.

n_down downsampling factor, ratio between interpolation rate and target rate.

nyquist_ratio low pass ﬁlter cutoff frequency relative to the nyquist frequency of the

smaller of the two sampling rates. Example values: E.g. 0.8, 0.9

n_irs length of impulse response (in samples at interpolation rate)

n_ringbuffer length of ringbuffer, in samples at source sampling rate

n_channels audio channels count

n_preﬁll Preﬁll the ringbuffer with this many zero frames in samples at source

sampling rate

562 CONTENTS

5.203.3 Member Function Documentation

5.203.3.1 void MHAFilter::polyphase_resampling_t::write (

mha_wave_t &signal )

Write signal to the ringbuffer.

Signal contained in signal is appended to the audio frames already present in the ringbuffer.

Parameters

signal input signal in original sampling rate

Exceptions

MHA_Error (p. 445)Raises exception if there is not enough room or if the number of

channels does not match.

5.203.3.2 void MHAFilter::polyphase_resampling_t::read (

mha_wave_t &signal )

Read resampled signal.

Will perform the resampling and remove no longer needed samples from the input buffer.

Parameters

signal buffer to write the resampled signal to.

Exceptions

MHA_Error (p. 445)Raises exception if there is not enough input signal or if the number

of channels is too high.

5.203.3.3 unsigned MHAFilter::polyphase_resampling_t::readable_frames ( ) const [inline]

Number of frames at target sampling rate that can be produced.

This method only checks for enough future samples present, therefore, this number can be

positive and a read operation can still fail if there are not enough past samples present to

perform the ﬁltering for the ﬁrst output sample. This could only happen if the constructor pa-

rameters n_ringbuffer or n_preﬁll have been chosen too small, because otherwise the method

read (p. 562) ensures that enough past samples are present to compute the next target sample.

5.203 MHAFilter::polyphase_resampling_t Class Reference 563

5.203.4 Member Data Documentation

5.203.4.1 unsigned MHAFilter::polyphase_resampling_t::upsampling_factor [private]

Integer upsampling factor.

Interpolation rate divided by source rate.

5.203.4.2 unsigned MHAFilter::polyphase_resampling_t::downsampling_factor [private]

Integer downsampling factor.

Interpolation rate divided by target rate.

5.203.4.3 unsigned MHAFilter::polyphase_resampling_t::now_index [private]

Index of "now" in the interpolated sampling rate.

Todo Index into what? What is the meaning of now?

5.203.4.4 bool MHAFilter::polyphase_resampling_t::underﬂow [private]

Set to true when an underﬂow has occurred.

When this is true, then the object can no longer be used. Underﬂows have to be avoided by

clients, e.g. by checking that enough readable_frames (p. 562) are present before calling read

(p. 562)

5.203.4.5 MHAWindow::hanning_t MHAFilter::polyphase_resampling_t::impulse_response

[private]

Contains the impulse response of the lowpass ﬁlter needed for anti-aliasing.

The impulse response is stored at the interpolation sampling rate. We use an instance of M←-

HAWindow::hanning_t (p. 840) here because we are limiting the sinc impulse response with

a Hanning window (otherwise the impulse response would extend indeﬁnitely into past and

future). And the samples inside an MHAWindow::hanning_t (p. 840) can be altered with ∗=,

which our constructor does.

5.203.4.6 MHASignal::ringbuffer_t MHAFilter::polyphase_resampling_t::ringbuffer [private]

Storage of input signal.

Part of the polyphase resampling optimization is that apart from the FIR impulse response,

nothing is stored at the interpolation rate, saving memory and computation cycles.

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

564 CONTENTS

5.204 MHAFilter::resampling_ﬁlter_t Class Reference

Hann shaped low pass ﬁlter for resampling.

Inheritance diagram for MHAFilter::resampling_ﬁlter_t:

MHAFilter::resampling

_filter_t

MHAFilter::fftfilter_t

Public Member Functions

•resampling_ﬁlter_t (unsigned int fftlen, unsigned int irslen, unsigned int channels, un-

signed int Nup, unsigned int Ndown, double fCutOff)

Constructor.

Static Public Member Functions

• static unsigned int fragsize_validator (unsigned int fftlen, unsigned int irslen)

Private Attributes

• unsigned int fragsize

5.204.1 Detailed Description

Hann shaped low pass ﬁlter for resampling.

This class uses FFT ﬁlter at upsampled rate.

5.204.2 Constructor & Destructor Documentation

5.204.2.1 MHAFilter::resampling_ﬁlter_t::resampling_ﬁlter_t (

unsigned int fftlen,

unsigned int irslen,

unsigned int channels,

unsigned int Nup,

unsigned int Ndown,

double fCutOff )

Constructor.

5.205 MHAFilter::smoothspec_t Class Reference 565

Parameters

fftlen FFT length.

irslen Length of ﬁlter.

channels Number of channels to be ﬁltered.

Nup Upsampling ratio.

Ndown Downsampling ratio.

fCutOff Cut off frequency (relative to lower Nyquist Frequency)

5.204.3 Member Function Documentation

5.204.3.1 unsigned int MHAFilter::resampling_ﬁlter_t::fragsize_validator (

unsigned int fftlen,

unsigned int irslen )[static]

5.204.4 Member Data Documentation

5.204.4.1 unsigned int MHAFilter::resampling_ﬁlter_t::fragsize [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.205 MHAFilter::smoothspec_t Class Reference

Smooth spectral gains, create a windowed impulse response.

Public Member Functions

•smoothspec_t (unsigned int fftlen, unsigned int nchannels, const MHAWindow←-

::base_t &window, bool minphase, bool linphase_asym=false)

Constructor.

• void smoothspec (const mha_spec_t &s_in, mha_spec_t &s_out)

Create a smoothed spectrum.

• void smoothspec (mha_spec_t &spec)

Create a smoothed spectrum (in place)

• void spec2ﬁr (const mha_spec_t &spec, mha_wave_t &ﬁr)

Return FIR coefﬁcients.

•∼smoothspec_t ()

566 CONTENTS

Private Member Functions

• void internal_ﬁr (const mha_spec_t &)

Private Attributes

• unsigned int fftlen

• unsigned int nchannels

•MHAWindow::base_t window

•MHASignal::waveform_t tmp_wave

•MHASignal::spectrum_t tmp_spec

•MHASignal::minphase_t ∗minphase

• bool _linphase_asym

•mha_fft_t fft

5.205.1 Detailed Description

Smooth spectral gains, create a windowed impulse response.

Spectral gains are smoothed by multiplicating the impulse response with a window function.

If a minimal phase is used, then the original phase is discarded and replaced by the minimal

phase function. In this case, the window is applied to the beginning of the inverse Fourier

transform of the input spectrum, and the remaining signal set to zero. If the original phase

is kept, the window is applied symmetrical arround zero, i.e. to the ﬁrst and last samples of

the inverse Fourier transform of the input spectrum. The spec2ﬁr() (p. 567) function creates a

causal impulse response by circular shifting the impulse response by half of the window length.

The signal dimensions of the arguments of smoothspec() (p. 567) must correspond to the FFT

length and number of channels provided in the constructor. The function spec2ﬁr() (p. 567)

can ﬁll signal structures with more than window length frames.

5.205.2 Constructor & Destructor Documentation

5.205.2.1 MHAFilter::smoothspec_t::smoothspec_t (

unsigned int fftlen,

unsigned int nchannels,

const MHAWindow::base_t &window,

bool minphase,

bool linphase_asym = false )

Constructor.

5.205 MHAFilter::smoothspec_t Class Reference 567

Parameters

fftlen FFT length of input spectrum (fftlen/2+1 bins)

nchannels Number of channels in input spectrum

window Window used for smoothing

minphase Use minimal phase (true) or original phase (false)

linphase_asym Keep phase, but apply full window at beginning of IRS

5.205.2.2 MHAFilter::smoothspec_t::∼smoothspec_t ( )

5.205.3 Member Function Documentation

5.205.3.1 void MHAFilter::smoothspec_t::smoothspec (

const mha_spec_t &s_in,

mha_spec_t &s_out )

Create a smoothed spectrum.

Parameters

s←-

_in

Input spectrum

Return values

s_out Output spectrum

5.205.3.2 void MHAFilter::smoothspec_t::smoothspec (

mha_spec_t &spec )[inline]

Create a smoothed spectrum (in place)

Parameters

spec Spectrum to be smoothed.

5.205.3.3 void MHAFilter::smoothspec_t::spec2ﬁr (

const mha_spec_t &spec,

mha_wave_t &ﬁr )

Return FIR coefﬁcients.

568 CONTENTS

Parameters

spec Input spectrum

Return values

ﬁr FIR coefﬁcients, minimum length is window length

5.205.3.4 void MHAFilter::smoothspec_t::internal_ﬁr (

const mha_spec_t &s_in )[private]

5.205.4 Member Data Documentation

5.205.4.1 unsigned int MHAFilter::smoothspec_t::fftlen [private]

5.205.4.2 unsigned int MHAFilter::smoothspec_t::nchannels [private]

5.205.4.3 MHAWindow::base_t MHAFilter::smoothspec_t::window [private]

5.205.4.4 MHASignal::waveform_t MHAFilter::smoothspec_t::tmp_wave [private]

5.205.4.5 MHASignal::spectrum_t MHAFilter::smoothspec_t::tmp_spec [private]

5.205.4.6 MHASignal::minphase_t∗MHAFilter::smoothspec_t::minphase [private]

5.205.4.7 bool MHAFilter::smoothspec_t::_linphase_asym [private]

5.205.4.8 mha_fft_t MHAFilter::smoothspec_t::fft [private]

The documentation for this class was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.206 MHAFilter::thirdoctave_analyzer_t Class Reference

Public Member Functions

•thirdoctave_analyzer_t (mhaconﬁg_t cfg)

•mha_wave_t ∗process (mha_wave_t ∗)

• unsigned int nbands ()

• unsigned int nchannels ()

• std::vector<mha_real_t >get_cf_hz ()

5.206 MHAFilter::thirdoctave_analyzer_t Class Reference 569

Static Public Member Functions

• static std::vector<mha_real_t >cf_generator (mhaconﬁg_t cfg)

• static std::vector<mha_real_t >bw_generator (mhaconﬁg_t cfg)

• static std::vector<mha_real_t >dup (std::vector<mha_real_t >,mhaconﬁg_t cfg)

Private Attributes

•mhaconﬁg_t cfg_

• std::vector<mha_real_t >cf

•MHAFilter::gamma_ﬂt_t fb

•MHASignal::waveform_t out_chunk

•MHASignal::waveform_t out_chunk_im

5.206.1 Constructor & Destructor Documentation

5.206.1.1 MHAFilter::thirdoctave_analyzer_t::thirdoctave_analyzer_t (

mhaconﬁg_t cfg )

5.206.2 Member Function Documentation

5.206.2.1 mha_wave_t ∗MHAFilter::thirdoctave_analyzer_t::process (

mha_wave_t ∗sIn )

5.206.2.2 unsigned int MHAFilter::thirdoctave_analyzer_t::nbands ( )

5.206.2.3 unsigned int MHAFilter::thirdoctave_analyzer_t::nchannels ( )

5.206.2.4 std::vector<mha_real_t >MHAFilter::thirdoctave_analyzer_t::get_cf_hz ( )

5.206.2.5 std::vector<mha_real_t >MHAFilter::thirdoctave_analyzer_t::cf_generator (

mhaconﬁg_t cfg )[static]

5.206.2.6 std::vector<mha_real_t >MHAFilter::thirdoctave_analyzer_t::bw_generator (

mhaconﬁg_t cfg )[static]

5.206.2.7 std::vector<mha_real_t >MHAFilter::thirdoctave_analyzer_t::dup (

std::vector<mha_real_t >vec,

mhaconﬁg_t cfg )[static]

5.206.3 Member Data Documentation

5.206.3.1 mhaconﬁg_t MHAFilter::thirdoctave_analyzer_t::cfg_ [private]

570 CONTENTS

5.206.3.2 std::vector<mha_real_t>MHAFilter::thirdoctave_analyzer_t::cf [private]

5.206.3.3 MHAFilter::gamma_ﬂt_t MHAFilter::thirdoctave_analyzer_t::fb [private]

5.206.3.4 MHASignal::waveform_t MHAFilter::thirdoctave_analyzer_t::out_chunk [private]

5.206.3.5 MHASignal::waveform_t MHAFilter::thirdoctave_analyzer_t::out_chunk_im

[private]

The documentation for this class was generated from the following ﬁles:

•complex_ﬁlter.h

•complex_ﬁlter.cpp

5.207 MHAFilter::transfer_function_t Struct Reference

a structure containing a source channel number, a target channel number, and an impulse

response.

Public Member Functions

•transfer_function_t ()

Default constructor for STL conformity.

•transfer_function_t (unsigned int source_channel_index, unsigned int target_←-

channel_index, const std::vector<ﬂoat >&impulse_response)

Data constructor.

• unsigned int partitions (unsigned int fragsize) const

for the given partition size, return the number of partitions of the impulse response.

• unsigned int non_empty_partitions (unsigned int fragsize) const

for the given partition size, return the number of non-empty partitions of the impulse response.

• bool isempty (unsigned int fragsize, unsigned int index) const

checks if the partition contains only zeros

Public Attributes

• unsigned int source_channel_index

Source audio channel index for this transfer function.

• unsigned int target_channel_index

Target audio channel index for this transfer function.

• std::vector<ﬂoat >impulse_response

Impulse response of transfer from source to target channel.

5.207 MHAFilter::transfer_function_t Struct Reference 571

5.207.1 Detailed Description

a structure containing a source channel number, a target channel number, and an impulse

response.

5.207.2 Constructor & Destructor Documentation

5.207.2.1 MHAFilter::transfer_function_t::transfer_function_t ( ) [inline]

Default constructor for STL conformity.

Not used.

5.207.2.2 MHAFilter::transfer_function_t::transfer_function_t (

unsigned int source_channel_index,

unsigned int target_channel_index,

const std::vector<ﬂoat >&impulse_response )

Data constructor.

Parameters

source_channel_index Source audio channel index for this transfer function

target_channel_index Target audio channel index for this transfer function

impulse_response Impulse response of transfer from source to target channel

5.207.3 Member Function Documentation

5.207.3.1 unsigned int MHAFilter::transfer_function_t::partitions (

unsigned int fragsize ) const [inline]

for the given partition size, return the number of partitions of the impulse response.

Parameters

fragsize partition size

Returns

number of partitions occupied by the impulse response

572 CONTENTS

5.207.3.2 unsigned int MHAFilter::transfer_function_t::non_empty_partitions (

unsigned int fragsize ) const [inline]

for the given partition size, return the number of non-empty partitions of the impulse response.

Parameters

fragsize partition size

Returns

the number of non-empty partitions of the impulse response, i.e. partitions containing

only zeros are not counted.

5.207.3.3 bool MHAFilter::transfer_function_t::isempty (

unsigned int fragsize,

unsigned int index ) const [inline]

checks if the partition contains only zeros

Parameters

fragsize partition size

index partition index

Returns

true when this partition of the impulse response contains only zeros.

5.207.4 Member Data Documentation

5.207.4.1 unsigned int MHAFilter::transfer_function_t::source_channel_index

Source audio channel index for this transfer function.

5.207.4.2 unsigned int MHAFilter::transfer_function_t::target_channel_index

Target audio channel index for this transfer function.

5.207.4.3 std::vector<ﬂoat>MHAFilter::transfer_function_t::impulse_response

Impulse response of transfer from source to target channel.

The documentation for this struct was generated from the following ﬁles:

•mha_ﬁlter.hh

•mha_ﬁlter.cpp

5.208 MHAFilter::transfer_matrix_t Struct Reference 573

5.208 MHAFilter::transfer_matrix_t Struct Reference

A sparse matrix of transfer function partitionss.

Inherits vector<transfer_function_t >.

Public Member Functions

• std::valarray<unsigned int >partitions (unsigned fragsize) const

Returns an array of the results of calling the partitions() (p. 573) method on every matrix

member.

• std::valarray<unsigned int >non_empty_partitions (unsigned int fragsize) const

Returns an array of the results of calling the non_empty_partitions() (p. 573) method on every

matrix member.

5.208.1 Detailed Description

A sparse matrix of transfer function partitionss.

Each matrix element knows its position in the matrix, so they can be stored as a vector.

5.208.2 Member Function Documentation

5.208.2.1 std::valarray<unsigned int>MHAFilter::transfer_matrix_t::partitions (

unsigned fragsize ) const [inline]

Returns an array of the results of calling the partitions() (p. 573) method on every matrix mem-

ber.

5.208.2.2 std::valarray<unsigned int>MHAFilter::transfer_matrix_t::non_empty_partitions (

unsigned int fragsize ) const [inline]

Returns an array of the results of calling the non_empty_partitions() (p. 573) method on every

matrix member.

The documentation for this struct was generated from the following ﬁle:

•mha_ﬁlter.hh

574 CONTENTS

5.209 MHAIOJack::io_jack_t Class Reference

Main class for JACK IO.

Inheritance diagram for MHAIOJack::io_jack_t:

MHAIOJack::io_jack_t

MHAParser::parser_t

MHAParser::base_t

MHAJack::client_t

Public Member Functions

•io_jack_t (unsigned int fragsize, ﬂoat samplerate,IOProcessEvent_t proc_event, void

∗proc_handle,IOStartedEvent_t start_event, void ∗start_handle,IOStoppedEvent←-

_t stop_event, void ∗stop_handle)

• void prepare (int, int)

Allocate buffers, activate JACK client and install internal ports.

• void release ()

Private Member Functions

• void reconnect_inports ()

Connect the input ports when connection variable is accessed.

• void reconnect_outports ()

Connect the output ports when connection variable is accessed.

• void get_physical_input_ports ()

• void get_physical_output_ports ()

• void get_all_input_ports ()

• void get_all_output_ports ()

• void get_delays_in ()

• void get_delays_out ()

• void read_get_cpu_load ()

• void read_get_xruns ()

• void read_get_scheduler ()

5.209 MHAIOJack::io_jack_t Class Reference 575

Private Attributes

• unsigned int fw_fragsize

• ﬂoat fw_samplerate

•MHAParser::string_t servername

•MHAParser::string_t clientname

•MHAParser::vstring_t connections_in

•MHAParser::vint_mon_t delays_in

•MHAParser::vstring_t connections_out

•MHAParser::vint_mon_t delays_out

•MHAParser::vstring_t portnames_in

•MHAParser::vstring_t portnames_out

•MHAParser::vstring_mon_t ports_in_physical

•MHAParser::vstring_mon_t ports_out_physical

•MHAParser::vstring_mon_t ports_in_all

•MHAParser::vstring_mon_t ports_out_all

•MHAParser::parser_t ports_parser

•MHAParser::ﬂoat_mon_t state_cpuload

•MHAParser::int_mon_t state_xruns

•MHAParser::int_mon_t state_priority

•MHAParser::string_mon_t state_scheduler

•MHAParser::parser_t state_parser

•MHAEvents::patchbay_t<io_jack_t >patchbay

Additional Inherited Members

5.209.1 Detailed Description

Main class for JACK IO.

This class registers a JACK client. JACK and framework states are managed by this class.

5.209.2 Constructor & Destructor Documentation

5.209.2.1 io_jack_t::io_jack_t (

unsigned int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

5.209.3 Member Function Documentation

5.209.3.1 void io_jack_t::prepare (

int nch_in,

int nch_out )

Allocate buffers, activate JACK client and install internal ports.

576 CONTENTS

5.209.3.2 void io_jack_t::release (

void )

5.209.3.3 void io_jack_t::reconnect_inports ( ) [private]

Connect the input ports when connection variable is accessed.

5.209.3.4 void io_jack_t::reconnect_outports ( ) [private]

Connect the output ports when connection variable is accessed.

5.209.3.5 void io_jack_t::get_physical_input_ports ( ) [private]

5.209.3.6 void io_jack_t::get_physical_output_ports ( ) [private]

5.209.3.7 void io_jack_t::get_all_input_ports ( ) [private]

5.209.3.8 void io_jack_t::get_all_output_ports ( ) [private]

5.209.3.9 void io_jack_t::get_delays_in ( ) [private]

5.209.3.10 void io_jack_t::get_delays_out ( ) [private]

5.209.3.11 void io_jack_t::read_get_cpu_load ( ) [private]

5.209.3.12 void io_jack_t::read_get_xruns ( ) [private]

5.209.3.13 void io_jack_t::read_get_scheduler ( ) [private]

5.209.4 Member Data Documentation

5.209.4.1 unsigned int MHAIOJack::io_jack_t::fw_fragsize [private]

5.209.4.2 ﬂoat MHAIOJack::io_jack_t::fw_samplerate [private]

5.209.4.3 MHAParser::string_t MHAIOJack::io_jack_t::servername [private]

5.209.4.4 MHAParser::string_t MHAIOJack::io_jack_t::clientname [private]

5.209.4.5 MHAParser::vstring_t MHAIOJack::io_jack_t::connections_in [private]

5.209.4.6 MHAParser::vint_mon_t MHAIOJack::io_jack_t::delays_in [private]

5.209 MHAIOJack::io_jack_t Class Reference 577

5.209.4.7 MHAParser::vstring_t MHAIOJack::io_jack_t::connections_out [private]

5.209.4.8 MHAParser::vint_mon_t MHAIOJack::io_jack_t::delays_out [private]

5.209.4.9 MHAParser::vstring_t MHAIOJack::io_jack_t::portnames_in [private]

5.209.4.10 MHAParser::vstring_t MHAIOJack::io_jack_t::portnames_out [private]

5.209.4.11 MHAParser::vstring_mon_t MHAIOJack::io_jack_t::ports_in_physical [private]

5.209.4.12 MHAParser::vstring_mon_t MHAIOJack::io_jack_t::ports_out_physical [private]

5.209.4.13 MHAParser::vstring_mon_t MHAIOJack::io_jack_t::ports_in_all [private]

5.209.4.14 MHAParser::vstring_mon_t MHAIOJack::io_jack_t::ports_out_all [private]

5.209.4.15 MHAParser::parser_t MHAIOJack::io_jack_t::ports_parser [private]

5.209.4.16 MHAParser::ﬂoat_mon_t MHAIOJack::io_jack_t::state_cpuload [private]

5.209.4.17 MHAParser::int_mon_t MHAIOJack::io_jack_t::state_xruns [private]

5.209.4.18 MHAParser::int_mon_t MHAIOJack::io_jack_t::state_priority [private]

5.209.4.19 MHAParser::string_mon_t MHAIOJack::io_jack_t::state_scheduler [private]

5.209.4.20 MHAParser::parser_t MHAIOJack::io_jack_t::state_parser [private]

5.209.4.21 MHAEvents::patchbay_t<io_jack_t>MHAIOJack::io_jack_t::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOJack.cpp

578 CONTENTS

5.210 MHAIOJackdb::io_jack_t Class Reference

Main class for JACK IO.

Inheritance diagram for MHAIOJackdb::io_jack_t:

MHAIOJackdb::io_jack_t

MHAParser::parser_t

MHAParser::base_t

MHAJack::client_t

Public Member Functions

•io_jack_t (unsigned int fragsize, ﬂoat samplerate,IOProcessEvent_t proc_event, void

∗proc_handle,IOStartedEvent_t start_event, void ∗start_handle,IOStoppedEvent←-

_t stop_event, void ∗stop_handle)

• void prepare (int, int)

Allocate buffers, activate JACK client and install internal ports.

• void release ()

• bool fail_on_async_jackerror () const

Private Member Functions

• int IOProcessEvent_inner (mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

• void reconnect_inports ()

Connect the input ports when connection variable is accessed.

• void reconnect_outports ()

Connect the output ports when connection variable is accessed.

• void get_physical_input_ports ()

• void get_physical_output_ports ()

• void get_all_input_ports ()

• void get_all_output_ports ()

• void read_get_cpu_load ()

• void read_get_xruns ()

• void read_get_scheduler ()

• void set_use_jack_transport ()

• void set_locate ()

5.210 MHAIOJackdb::io_jack_t Class Reference 579

Static Private Member Functions

• static int IOProcessEvent_inner (void ∗handle, mha_wave_t ∗sIn, mha_wave_t ∗∗s←-

Out)

Private Attributes

•IOProcessEvent_t proc_event

• void ∗proc_handle

• unsigned int mha_fragsize

• ﬂoat mha_samplerate

• unsigned int fragsize_ratio

•MHAParser::string_t servername

•MHAParser::string_t clientname

•MHAParser::vstring_t connections_in

•MHAParser::vstring_t connections_out

•MHAParser::vstring_t portnames_in

•MHAParser::vstring_t portnames_out

•MHAParser::bool_t fail_on_async_jackerr

•MHAParser::bool_t use_jack_transport

•MHAParser::ﬂoat_t locate

•MHAParser::ﬂoat_mon_t server_srate

•MHAParser::int_mon_t server_fragsize

•MHAParser::vstring_mon_t ports_in_physical

•MHAParser::vstring_mon_t ports_out_physical

•MHAParser::vstring_mon_t ports_in_all

•MHAParser::vstring_mon_t ports_out_all

•MHAParser::parser_t ports_parser

•MHAParser::ﬂoat_mon_t state_cpuload

•MHAParser::int_mon_t state_xruns

•MHAParser::int_mon_t state_priority

•MHAParser::string_mon_t state_scheduler

•MHAParser::parser_t state_parser

•MHASignal::waveform_t ∗pwinner_out

•MHAEvents::patchbay_t<io_jack_t >patchbay

Additional Inherited Members

5.210.1 Detailed Description

Main class for JACK IO.

This class registers a JACK client. JACK and framework states are managed by this class.

580 CONTENTS

5.210.2 Constructor & Destructor Documentation

5.210.2.1 io_jack_t::io_jack_t (

unsigned int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )

5.210.3 Member Function Documentation

5.210.3.1 void io_jack_t::prepare (

int nch_in,

int nch_out )

Allocate buffers, activate JACK client and install internal ports.

5.210.3.2 void io_jack_t::release (

void )

5.210.3.3 bool MHAIOJackdb::io_jack_t::fail_on_async_jackerror ( ) const [inline]

5.210.3.4 int io_jack_t::IOProcessEvent_inner (

void ∗handle,

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[static],[private]

5.210.3.5 int io_jack_t::IOProcessEvent_inner (

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[private]

5.210.3.6 void io_jack_t::reconnect_inports ( ) [private]

Connect the input ports when connection variable is accessed.

5.210.3.7 void io_jack_t::reconnect_outports ( ) [private]

Connect the output ports when connection variable is accessed.

5.210 MHAIOJackdb::io_jack_t Class Reference 581

5.210.3.8 void io_jack_t::get_physical_input_ports ( ) [private]

5.210.3.9 void io_jack_t::get_physical_output_ports ( ) [private]

5.210.3.10 void io_jack_t::get_all_input_ports ( ) [private]

5.210.3.11 void io_jack_t::get_all_output_ports ( ) [private]

5.210.3.12 void io_jack_t::read_get_cpu_load ( ) [private]

5.210.3.13 void io_jack_t::read_get_xruns ( ) [private]

5.210.3.14 void io_jack_t::read_get_scheduler ( ) [private]

5.210.3.15 void io_jack_t::set_use_jack_transport ( ) [private]

5.210.3.16 void io_jack_t::set_locate ( ) [private]

5.210.4 Member Data Documentation

5.210.4.1 IOProcessEvent_t MHAIOJackdb::io_jack_t::proc_event [private]

5.210.4.2 void∗MHAIOJackdb::io_jack_t::proc_handle [private]

5.210.4.3 unsigned int MHAIOJackdb::io_jack_t::mha_fragsize [private]

5.210.4.4 ﬂoat MHAIOJackdb::io_jack_t::mha_samplerate [private]

5.210.4.5 unsigned int MHAIOJackdb::io_jack_t::fragsize_ratio [private]

5.210.4.6 MHAParser::string_t MHAIOJackdb::io_jack_t::servername [private]

5.210.4.7 MHAParser::string_t MHAIOJackdb::io_jack_t::clientname [private]

5.210.4.8 MHAParser::vstring_t MHAIOJackdb::io_jack_t::connections_in [private]

5.210.4.9 MHAParser::vstring_t MHAIOJackdb::io_jack_t::connections_out [private]

5.210.4.10 MHAParser::vstring_t MHAIOJackdb::io_jack_t::portnames_in [private]

5.210.4.11 MHAParser::vstring_t MHAIOJackdb::io_jack_t::portnames_out [private]

5.210.4.12 MHAParser::bool_t MHAIOJackdb::io_jack_t::fail_on_async_jackerr [private]

582 CONTENTS

5.210.4.13 MHAParser::bool_t MHAIOJackdb::io_jack_t::use_jack_transport [private]

5.210.4.14 MHAParser::ﬂoat_t MHAIOJackdb::io_jack_t::locate [private]

5.210.4.15 MHAParser::ﬂoat_mon_t MHAIOJackdb::io_jack_t::server_srate [private]

5.210.4.16 MHAParser::int_mon_t MHAIOJackdb::io_jack_t::server_fragsize [private]

5.210.4.17 MHAParser::vstring_mon_t MHAIOJackdb::io_jack_t::ports_in_physical [private]

5.210.4.18 MHAParser::vstring_mon_t MHAIOJackdb::io_jack_t::ports_out_physical

[private]

5.210.4.19 MHAParser::vstring_mon_t MHAIOJackdb::io_jack_t::ports_in_all [private]

5.210.4.20 MHAParser::vstring_mon_t MHAIOJackdb::io_jack_t::ports_out_all [private]

5.210.4.21 MHAParser::parser_t MHAIOJackdb::io_jack_t::ports_parser [private]

5.210.4.22 MHAParser::ﬂoat_mon_t MHAIOJackdb::io_jack_t::state_cpuload [private]

5.210.4.23 MHAParser::int_mon_t MHAIOJackdb::io_jack_t::state_xruns [private]

5.210.4.24 MHAParser::int_mon_t MHAIOJackdb::io_jack_t::state_priority [private]

5.210.4.25 MHAParser::string_mon_t MHAIOJackdb::io_jack_t::state_scheduler [private]

5.210.4.26 MHAParser::parser_t MHAIOJackdb::io_jack_t::state_parser [private]

5.210.4.27 MHASignal::waveform_t∗MHAIOJackdb::io_jack_t::pwinner_out [private]

5.210.4.28 MHAEvents::patchbay_t<io_jack_t>MHAIOJackdb::io_jack_t::patchbay

[private]

The documentation for this class was generated from the following ﬁle:

•MHAIOJackdb.cpp

5.211 MHAIOPortAudio::device_info_t Class Reference 583

5.211 MHAIOPortAudio::device_info_t Class Reference

Inheritance diagram for MHAIOPortAudio::device_info_t:

MHAIOPortAudio::device

_info_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•device_info_t ()

• void ﬁll_info ()

Public Attributes

•MHAParser::int_mon_t numDevices

•MHAParser::vint_mon_t structVersion

•MHAParser::vstring_mon_t name

•MHAParser::vint_mon_t hostApi

•MHAParser::vint_mon_t maxInputChannels

•MHAParser::vint_mon_t maxOutputChannels

•MHAParser::vﬂoat_mon_t defaultLowInputLatency

•MHAParser::vﬂoat_mon_t defaultLowOutputLatency

•MHAParser::vﬂoat_mon_t defaultHighInputLatency

•MHAParser::vﬂoat_mon_t defaultHighOutputLatency

•MHAParser::vﬂoat_mon_t defaultSampleRate

584 CONTENTS

Additional Inherited Members

5.211.1 Constructor & Destructor Documentation

5.211.1.1 MHAIOPortAudio::device_info_t::device_info_t ( ) [inline]

5.211.2 Member Function Documentation

5.211.2.1 void MHAIOPortAudio::device_info_t::ﬁll_info ( ) [inline]

5.211.3 Member Data Documentation

5.211.3.1 MHAParser::int_mon_t MHAIOPortAudio::device_info_t::numDevices

5.211.3.2 MHAParser::vint_mon_t MHAIOPortAudio::device_info_t::structVersion

5.211.3.3 MHAParser::vstring_mon_t MHAIOPortAudio::device_info_t::name

5.211.3.4 MHAParser::vint_mon_t MHAIOPortAudio::device_info_t::hostApi

5.211.3.5 MHAParser::vint_mon_t MHAIOPortAudio::device_info_t::maxInputChannels

5.211.3.6 MHAParser::vint_mon_t MHAIOPortAudio::device_info_t::maxOutputChannels

5.211.3.7 MHAParser::vﬂoat_mon_t MHAIOPortAudio::device_info_t::defaultLowInputLatency

5.211.3.8 MHAParser::vﬂoat_mon_t MHAIOPortAudio::device_info_t::defaultLowOutputLatency

5.211.3.9 MHAParser::vﬂoat_mon_t MHAIOPortAudio::device_info_t::defaultHighInputLatency

5.211.3.10 MHAParser::vﬂoat_mon_t MHAIOPortAudio::device_info_t::defaultHighOutputLatency

5.211.3.11 MHAParser::vﬂoat_mon_t MHAIOPortAudio::device_info_t::defaultSampleRate

The documentation for this class was generated from the following ﬁle:

•MHAIOPortAudio.cpp

5.212 MHAIOPortAudio::io_portaudio_t Class Reference 585

5.212 MHAIOPortAudio::io_portaudio_t Class Reference

Main class for Portaudio sound IO.

Inheritance diagram for MHAIOPortAudio::io_portaudio_t:

MHAIOPortAudio::io

_portaudio_t

MHAParser::parser_t

MHAParser::base_t

Public Member Functions

•io_portaudio_t (unsigned int fragsize, ﬂoat samplerate,IOProcessEvent_t proc_←-

event, void ∗proc_handle,IOStartedEvent_t start_event, void ∗start_handle,IO←-

StoppedEvent_t stop_event, void ∗stop_handle)

• void device_name_updated ()

• void device_index_updated ()

•∼io_portaudio_t ()

• void cmd_prepare (int, int)

• void cmd_start ()

• void cmd_stop ()

• void cmd_release ()

• int portaudio_callback (const void ∗input, void ∗output, unsigned long frame_count,

const PaStreamCallbackTimeInfo ∗time_info, PaStreamCallbackFlags status_ﬂags)

Private Attributes

•device_info_t device_info

•MHASignal::waveform_t ∗s_in

•mha_wave_t ∗s_out

• ﬂoat samplerate

• unsigned int nchannels_out

• unsigned int nchannels_in

586 CONTENTS

• unsigned int fragsize

•IOProcessEvent_t proc_event

• void ∗proc_handle

•IOStartedEvent_t start_event

• void ∗start_handle

•IOStoppedEvent_t stop_event

• void ∗stop_handle

• PaStream ∗portaudio_stream

•MHAParser::string_t device_name

•MHAParser::int_t device_index

•MHAEvents::patchbay_t<io_portaudio_t >patchbay

Additional Inherited Members

5.212.1 Detailed Description

Main class for Portaudio sound IO.

5.212.2 Constructor & Destructor Documentation

5.212.2.1 MHAIOPortAudio::io_portaudio_t::io_portaudio_t (

unsigned int fragsize,

ﬂoat samplerate,

IOProcessEvent_t proc_event,

void ∗proc_handle,

IOStartedEvent_t start_event,

void ∗start_handle,

IOStoppedEvent_t stop_event,

void ∗stop_handle )[inline]

5.212.2.2 MHAIOPortAudio::io_portaudio_t::∼io_portaudio_t ( ) [inline]

5.212.3 Member Function Documentation

5.212.3.1 void MHAIOPortAudio::io_portaudio_t::device_name_updated ( ) [inline]

5.212.3.2 void MHAIOPortAudio::io_portaudio_t::device_index_updated ( ) [inline]

5.212.3.3 void MHAIOPortAudio::io_portaudio_t::cmd_prepare (

int nchannels_in,

int nchannels_out )

5.212.3.4 void MHAIOPortAudio::io_portaudio_t::cmd_start ( )

5.212 MHAIOPortAudio::io_portaudio_t Class Reference 587

5.212.3.5 void MHAIOPortAudio::io_portaudio_t::cmd_stop ( )

5.212.3.6 void MHAIOPortAudio::io_portaudio_t::cmd_release ( )

5.212.3.7 int MHAIOPortAudio::io_portaudio_t::portaudio_callback (

const void ∗input,

void ∗output,

unsigned long frame_count,

const PaStreamCallbackTimeInfo ∗time_info,

PaStreamCallbackFlags status_ﬂags )

5.212.4 Member Data Documentation

5.212.4.1 device_info_t MHAIOPortAudio::io_portaudio_t::device_info [private]

5.212.4.2 MHASignal::waveform_t∗MHAIOPortAudio::io_portaudio_t::s_in [private]

5.212.4.3 mha_wave_t∗MHAIOPortAudio::io_portaudio_t::s_out [private]

5.212.4.4 ﬂoat MHAIOPortAudio::io_portaudio_t::samplerate [private]

5.212.4.5 unsigned int MHAIOPortAudio::io_portaudio_t::nchannels_out [private]

5.212.4.6 unsigned int MHAIOPortAudio::io_portaudio_t::nchannels_in [private]

5.212.4.7 unsigned int MHAIOPortAudio::io_portaudio_t::fragsize [private]

5.212.4.8 IOProcessEvent_t MHAIOPortAudio::io_portaudio_t::proc_event [private]

5.212.4.9 void∗MHAIOPortAudio::io_portaudio_t::proc_handle [private]

5.212.4.10 IOStartedEvent_t MHAIOPortAudio::io_portaudio_t::start_event [private]

5.212.4.11 void∗MHAIOPortAudio::io_portaudio_t::start_handle [private]

5.212.4.12 IOStoppedEvent_t MHAIOPortAudio::io_portaudio_t::stop_event [private]

5.212.4.13 void∗MHAIOPortAudio::io_portaudio_t::stop_handle [private]

5.212.4.14 PaStream∗MHAIOPortAudio::io_portaudio_t::portaudio_stream [private]

5.212.4.15 MHAParser::string_t MHAIOPortAudio::io_portaudio_t::device_name [private]

5.212.4.16 MHAParser::int_t MHAIOPortAudio::io_portaudio_t::device_index [private]

5.212.4.17 MHAEvents::patchbay_t<io_portaudio_t>MHAIOPortAudio::io_portaudio_t←-

::patchbay [private]

The documentation for this class was generated from the following ﬁle:

•MHAIOPortAudio.cpp

588 CONTENTS

5.213 MHAJack::client_avg_t Class Reference

Generic JACK client for averaging a system response across time.

Inheritance diagram for MHAJack::client_avg_t:

MHAJack::client_avg_t

MHAJack::client_t

Public Member Functions

•client_avg_t (const std::string &name, const unsigned int &nrep_)

Constructor for averaging client.

• void io (mha_wave_t ∗s_out,mha_wave_t ∗s_in, const std::vector<std::string

>&p_out, const std::vector<std::string >&p_in, ﬂoat ∗srate=NULL, unsigned int

∗fragsize=NULL)

Recording function.

Private Member Functions

• void proc (mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

• void IOStoppedEvent ()

Static Private Member Functions

• static int proc (void ∗handle, mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

• static void IOStoppedEvent (void ∗handle, int proc_err, int io_err)

Private Attributes

• bool b_stopped

• unsigned int pos

•mha_wave_t ∗sn_in

•mha_wave_t ∗sn_out

• std::string name

•MHASignal::waveform_t ∗frag_out

• const unsigned int nrep

• unsigned int n

• bool b_ready

5.213 MHAJack::client_avg_t Class Reference 589

Additional Inherited Members

5.213.1 Detailed Description

Generic JACK client for averaging a system response across time.

5.213.2 Constructor & Destructor Documentation

5.213.2.1 MHAJack::client_avg_t::client_avg_t (

const std::string & name_,

const unsigned int & nrep_ )

Constructor for averaging client.

Parameters

name←-

Name of JACK client

nrep←-

Number of repetitions

5.213.3 Member Function Documentation

5.213.3.1 void MHAJack::client_avg_t::io (

mha_wave_t ∗is_out,

mha_wave_t ∗is_in,

const std::vector<std::string >&p_out,

const std::vector<std::string >&p_in,

ﬂoat ∗srate = NULL,

unsigned int ∗fragsize = NULL )

Recording function.

long-description

Parameters

is_out Input (test) signal, which will be repeated

is_in System response (averaged, same length as input required)

p_out Ports to play back the test signal

p_in Ports to record from the system response

srate Pointer to sampling rate variable, will be ﬁlled with server sampling rate

fragsize Pointer to fragment size variable, will be ﬁlled with server fragment size

590 CONTENTS

5.213.3.2 int MHAJack::client_avg_t::proc (

void ∗handle,

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[static],[private]

5.213.3.3 void MHAJack::client_avg_t::IOStoppedEvent (

void ∗handle,

int proc_err,

int io_err )[static],[private]

5.213.3.4 void MHAJack::client_avg_t::proc (

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[private]

5.213.3.5 void MHAJack::client_avg_t::IOStoppedEvent ( ) [private]

5.213.4 Member Data Documentation

5.213.4.1 bool MHAJack::client_avg_t::b_stopped [private]

5.213.4.2 unsigned int MHAJack::client_avg_t::pos [private]

5.213.4.3 mha_wave_t∗MHAJack::client_avg_t::sn_in [private]

5.213.4.4 mha_wave_t∗MHAJack::client_avg_t::sn_out [private]

5.213.4.5 std::string MHAJack::client_avg_t::name [private]

5.213.4.6 MHASignal::waveform_t∗MHAJack::client_avg_t::frag_out [private]

5.213.4.7 const unsigned int MHAJack::client_avg_t::nrep [private]

5.213.4.8 unsigned int MHAJack::client_avg_t::n [private]

5.213.4.9 bool MHAJack::client_avg_t::b_ready [private]

The documentation for this class was generated from the following ﬁles:

•mhajack.h

•mhajack.cpp

5.214 MHAJack::client_noncont_t Class Reference 591

5.214 MHAJack::client_noncont_t Class Reference

Generic client for synchronous playback and recording of waveform fragments.

Inheritance diagram for MHAJack::client_noncont_t:

MHAJack::client_noncont_t

MHAJack::client_t

Public Member Functions

•client_noncont_t (const std::string &name, bool use_jack_transport=false)

• void io (mha_wave_t ∗s_out,mha_wave_t ∗s_in, const std::vector<std::string

>&p_out, const std::vector<std::string >&p_in, ﬂoat ∗srate=NULL, unsigned int

∗fragsize=NULL)

Private Member Functions

• void proc (mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

• void IOStoppedEvent ()

Static Private Member Functions

• static int proc (void ∗handle, mha_wave_t ∗sIn, mha_wave_t ∗∗sOut)

• static void IOStoppedEvent (void ∗handle, int proc_err, int io_err)

Private Attributes

• bool b_stopped

• unsigned int pos

•mha_wave_t ∗sn_in

•mha_wave_t ∗sn_out

• std::string name

•MHASignal::waveform_t ∗frag_out

592 CONTENTS

Additional Inherited Members

5.214.1 Detailed Description

Generic client for synchronous playback and recording of waveform fragments.

5.214.2 Constructor & Destructor Documentation

5.214.2.1 MHAJack::client_noncont_t::client_noncont_t (

const std::string & name,

bool use_jack_transport = false )

5.214.3 Member Function Documentation

5.214.3.1 void MHAJack::client_noncont_t::io (

mha_wave_t ∗s_out,

mha_wave_t ∗s_in,

const std::vector<std::string >&p_out,

const std::vector<std::string >&p_in,

ﬂoat ∗srate = NULL,

unsigned int ∗fragsize = NULL )

5.214.3.2 int MHAJack::client_noncont_t::proc (

void ∗handle,

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[static],[private]

5.214.3.3 void MHAJack::client_noncont_t::IOStoppedEvent (

void ∗handle,

int proc_err,

int io_err )[static],[private]

5.214.3.4 void MHAJack::client_noncont_t::proc (

mha_wave_t ∗sIn,

mha_wave_t ∗∗ sOut )[private]

5.214.3.5 void MHAJack::client_noncont_t::IOStoppedEvent ( ) [private]

5.214.4 Member Data Documentation

5.214.4.1 bool MHAJack::client_noncont_t::b_stopped [private]

5.214.4.2 unsigned int MHAJack::client_noncont_t::pos [private]

5.215 MHAJack::client_t Class Reference 593

5.214.4.3 mha_wave_t∗MHAJack::client_noncont_t::sn_in [private]

5.214.4.4 mha_wave_t∗MHAJack::client_noncont_t::sn_out [private]

5.214.4.5 std::string MHAJack::client_noncont_t::name [private]

5.214.4.6 MHASignal::waveform_t∗MHAJack::client_noncont_t::frag_out [private]

The documentation for this class was generated from the following ﬁles:

•mhajack.h

•mhajack.cpp

5.215 MHAJack::client_t Class Reference

Generic asynchronous JACK client.

Inheritance diagram for MHAJack::client_t:

MHAJack::client_t

MHAIOJack::io_jack_t

MHAIOJackdb::io_jack_t

MHAJack::client_avg_t

MHAJack::client_noncont_t

Public Member Functions

•client_t (IOProcessEvent_t proc_event, void ∗proc_handle=NULL, IOStartedEvent←-

_t start_event=NULL, void ∗start_handle=NULL, IOStoppedEvent_t stop_event=N←-

ULL, void ∗stop_handle=NULL, bool use_jack_transport=false)

• void prepare (const std::string &client_name, const unsigned int &nchannels_in, const

unsigned int &nchannels_out)

Allocate buffers, activate JACK client and install internal ports.

594 CONTENTS

• void prepare (const std::string &server_name, const std::string &client_name, const un-

signed int &nchannels_in, const unsigned int &nchannels_out)

Allocate buffers, ports, and activates JACK client.

• void release ()

Remove JACK client and deallocate internal ports and buffers.

• void start (bool fail_on_async_jack_error=true)

• void stop ()

• void connect_input (const std::vector<std::string >&)

Connect the input ports when connection variable is accessed.

• void connect_output (const std::vector<std::string >&)

Connect the output ports when connection variable is accessed.

• unsigned int get_fragsize () const

• ﬂoat get_srate () const

• unsigned long get_xruns ()

• unsigned long get_xruns_reset ()

• std::string str_error (int err)

• void get_ports (std::vector<std::string >&, unsigned long jack_ﬂags)

Get a list of Jack ports.

• std::vector<std::string >get_my_input_ports ()

• std::vector<std::string >get_my_output_ports ()

• void set_input_portnames (const std::vector<std::string >&)

• void set_output_portnames (const std::vector<std::string >&)

• ﬂoat get_cpu_load ()

• void set_use_jack_transport (bool ut)

Protected Attributes

• jack_client_t ∗jc

Private Member Functions

• void prepare_impl (const char ∗server_name, const char ∗client_name, const unsigned

int &nchannels_in, const unsigned int &nchannels_out)

Allocate buffers, activate JACK client and allocates jack ports Registers the jack client with the

given server and activates it.

• void internal_start ()

• void internal_stop ()

• void stopped (int, int)

• int jack_proc_cb (jack_nframes_t)

This is the main processing callback.

• int jack_xrun_cb ()

Static Private Member Functions

• static int jack_proc_cb (jack_nframes_t, void ∗)

• static int jack_xrun_cb (void ∗)

5.215 MHAJack::client_t Class Reference 595

Private Attributes

• unsigned long num_xruns

• unsigned int fragsize

• ﬂoat samplerate

• unsigned int nchannels_in

• unsigned int nchannels_out

•IOProcessEvent_t proc_event

• void ∗proc_handle

•IOStartedEvent_t start_event

• void ∗start_handle

•IOStoppedEvent_t stop_event

• void ∗stop_handle

•MHASignal::waveform_t ∗s_in

•mha_wave_t ∗s_out

•MHAJack::port_t ∗∗ inch

•MHAJack::port_t ∗∗ outch

• unsigned int ﬂags

• bool b_prepared

• bool use_jack_transport

• jack_transport_state_t jstate_prev

• std::vector<std::string >input_portnames

• std::vector<std::string >output_portnames

• bool fail_on_async_jackerror

5.215.1 Detailed Description

Generic asynchronous JACK client.

5.215.2 Constructor & Destructor Documentation

5.215.2.1 MHAJack::client_t::client_t (

IOProcessEvent_t proc_event,

void ∗proc_handle = NULL,

IOStartedEvent_t start_event = NULL,

void ∗start_handle = NULL,

IOStoppedEvent_t stop_event = NULL,

void ∗stop_handle = NULL,

bool use_jack_transport = false )

5.215.3 Member Function Documentation

5.215.3.1 void MHAJack::client_t::prepare (

const std::string & client_name,

const unsigned int & nch_in,

const unsigned int & nch_out )

Allocate buffers, activate JACK client and install internal ports.

Registers the jack client with the default jack server and activates it.

596 CONTENTS

Parameters

client_name Name of this jack client

nch_in Input ports to register

nch_out Output ports to register

5.215.3.2 void MHAJack::client_t::prepare (

const std::string & server_name,

const std::string & client_name,

const unsigned int & nch_in,

const unsigned int & nch_out )

Allocate buffers, ports, and activates JACK client.

Registers the jack client with speciﬁed jack server and activates it.

Parameters

server_name Name of the jack server to register with

client_name Name of this jack client

nch_in Input ports to register

nch_out Output ports to register

5.215.3.3 void MHAJack::client_t::release (

void )

Remove JACK client and deallocate internal ports and buffers.

5.215.3.4 void MHAJack::client_t::start (

bool fail_on_async_jack_error = true )

5.215.3.5 void MHAJack::client_t::stop ( )

5.215.3.6 void MHAJack::client_t::connect_input (

const std::vector<std::string >&con )

Connect the input ports when connection variable is accessed.

5.215.3.7 void MHAJack::client_t::connect_output (

const std::vector<std::string >&con )

Connect the output ports when connection variable is accessed.

5.215 MHAJack::client_t Class Reference 597

5.215.3.8 unsigned int MHAJack::client_t::get_fragsize ( ) const [inline]

5.215.3.9 ﬂoat MHAJack::client_t::get_srate ( ) const [inline]

5.215.3.10 unsigned long MHAJack::client_t::get_xruns ( ) [inline]

5.215.3.11 unsigned long MHAJack::client_t::get_xruns_reset ( )

5.215.3.12 std::string MHAJack::client_t::str_error (

int err )

5.215.3.13 void MHAJack::client_t::get_ports (

std::vector<std::string >&res,

unsigned long jack_ﬂags )

Get a list of Jack ports.

Parameters

res Result string vector

jack_ﬂags Jack port ﬂags (JackPortInput etc.)

5.215.3.14 std::vector<std::string >MHAJack::client_t::get_my_input_ports ( )

5.215.3.15 std::vector<std::string >MHAJack::client_t::get_my_output_ports ( )

5.215.3.16 void MHAJack::client_t::set_input_portnames (

const std::vector<std::string >&names )

5.215.3.17 void MHAJack::client_t::set_output_portnames (

const std::vector<std::string >&names )

5.215.3.18 ﬂoat MHAJack::client_t::get_cpu_load ( )

5.215.3.19 void MHAJack::client_t::set_use_jack_transport (

bool ut )[inline]

5.215.3.20 void MHAJack::client_t::prepare_impl (

const char ∗server_name,

const char ∗client_name,

const unsigned int & nch_in,

const unsigned int & nch_out )[private]

Allocate buffers, activate JACK client and allocates jack ports Registers the jack client with the

given server and activates it.

598 CONTENTS

Parameters

server_name Name of the jack server to register with

client_name Name of this jack client

nch_in Input ports to register

nch_out Output ports to register

5.215.3.21 void MHAJack::client_t::internal_start ( ) [private]

5.215.3.22 void MHAJack::client_t::internal_stop ( ) [private]

5.215.3.23 void MHAJack::client_t::stopped (

int proc_err,

int io_err )[private]

5.215.3.24 int MHAJack::client_t::jack_proc_cb (

jack_nframes_t n,

void ∗h)[static],[private]

5.215.3.25 int MHAJack::client_t::jack_proc_cb (

jack_nframes_t n)[private]

This is the main processing callback.

Here happens double buffering and downsampling.

5.215.3.26 int MHAJack::client_t::jack_xrun_cb (

void ∗h)[static],[private]

5.215.3.27 int MHAJack::client_t::jack_xrun_cb ( ) [inline],[private]

5.215.4 Member Data Documentation

5.215.4.1 unsigned long MHAJack::client_t::num_xruns [private]

5.215.4.2 unsigned int MHAJack::client_t::fragsize [private]

5.215.4.3 ﬂoat MHAJack::client_t::samplerate [private]

5.215.4.4 unsigned int MHAJack::client_t::nchannels_in [private]

5.215.4.5 unsigned int MHAJack::client_t::nchannels_out [private]

5.215.4.6 IOProcessEvent_t MHAJack::client_t::proc_event [private]

5.216 MHAJack::port_t Class Reference 599

5.215.4.7 void∗MHAJack::client_t::proc_handle [private]

5.215.4.8 IOStartedEvent_t MHAJack::client_t::start_event [private]

5.215.4.9 void∗MHAJack::client_t::start_handle [private]

5.215.4.10 IOStoppedEvent_t MHAJack::client_t::stop_event [private]

5.215.4.11 void∗MHAJack::client_t::stop_handle [private]

5.215.4.12 MHASignal::waveform_t∗MHAJack::client_t::s_in [private]

5.215.4.13 mha_wave_t∗MHAJack::client_t::s_out [private]

5.215.4.14 MHAJack::port_t∗∗ MHAJack::client_t::inch [private]

5.215.4.15 MHAJack::port_t∗∗ MHAJack::client_t::outch [private]

5.215.4.16 jack_client_t∗MHAJack::client_t::jc [protected]

5.215.4.17 unsigned int MHAJack::client_t::ﬂags [private]

5.215.4.18 bool MHAJack::client_t::b_prepared [private]

5.215.4.19 bool MHAJack::client_t::use_jack_transport [private]

5.215.4.20 jack_transport_state_t MHAJack::client_t::jstate_prev [private]

5.215.4.21 std::vector<std::string>MHAJack::client_t::input_portnames [private]

5.215.4.22 std::vector<std::string>MHAJack::client_t::output_portnames [private]

5.215.4.23 bool MHAJack::client_t::fail_on_async_jackerror [private]

The documentation for this class was generated from the following ﬁles:

•mhajack.h

•mhajack.cpp

5.216 MHAJack::port_t Class Reference

Class for one channel/port.

600 CONTENTS

Public Types

Public Member Functions

•port_t (jack_client_t ∗jc,dir_t dir, int id)

•port_t (jack_client_t ∗jc,dir_t dir, const std::string &id)

Constructor to create port with speciﬁc name.

•∼port_t ()

• void read (mha_wave_t ∗s, unsigned int ch)

• void write (mha_wave_t ∗s, unsigned int ch)

• void mute (unsigned int n)

• void connect_to (const char ∗pn)

• const char ∗get_short_name ()

Return the port name.

Private Attributes

•dir_t dir_type

• jack_port_t ∗port

• jack_default_audio_sample_t ∗iob

• jack_client_t ∗jc

5.216.1 Detailed Description

Class for one channel/port.

This class represents one JACK port. Double buffering for asynchronous process callbacks is

managed by this class.

5.216.2 Member Enumeration Documentation

5.216.2.1 enum MHAJack::port_t::dir_t

Enumerator

input

output

5.216.3 Constructor & Destructor Documentation

5.216.3.1 MHAJack::port_t::port_t (

jack_client_t ∗jc,

dir_t dir,

int id )

5.216 MHAJack::port_t Class Reference 601

Parameters

jc JACK client.

dir Direction (input/output).

id Number in port name (starting with 1).

5.216.3.2 MHAJack::port_t::port_t (

jack_client_t ∗jc,

dir_t dir,

const std::string & id )

Constructor to create port with speciﬁc name.

Parameters

jc JACK client.

dir Direction (input/output).

id Port name.

5.216.3.3 MHAJack::port_t::∼port_t ( )

5.216.4 Member Function Documentation

5.216.4.1 void MHAJack::port_t::read (

mha_wave_t ∗s,

unsigned int ch )

Parameters

sSignal structure to store the audio data.

ch Channel number in audio data structure to be used.

5.216.4.2 void MHAJack::port_t::write (

mha_wave_t ∗s,

unsigned int ch )

Parameters

sSignal structure from which the audio data is read.

ch Channel number in audio data structure to be used.

5.216.4.3 void MHAJack::port_t::mute (

unsigned int n)

602 CONTENTS

Parameters

nNumber of samples to be muted (must be the same as reported by Jack processing

callback).

5.216.4.4 void MHAJack::port_t::connect_to (

const char ∗pn )

Parameters

pn Port name to connect to

5.216.4.5 const char ∗MHAJack::port_t::get_short_name ( )

Return the port name.

5.216.5 Member Data Documentation

5.216.5.1 dir_t MHAJack::port_t::dir_type [private]

5.216.5.2 jack_port_t∗MHAJack::port_t::port [private]

5.216.5.3 jack_default_audio_sample_t∗MHAJack::port_t::iob [private]

5.216.5.4 jack_client_t∗MHAJack::port_t::jc [private]

The documentation for this class was generated from the following ﬁles:

•mhajack.h

•mhajack.cpp

5.217 MHAKernel::algo_comm_class_t Class Reference 603

5.217 MHAKernel::algo_comm_class_t Class Reference

Inheritance diagram for MHAKernel::algo_comm_class_t:

MHAKernel::algo_comm

_class_t

mhaplug_cfg_t

MHAPlugin_Split::splitted

_part_t

plug_t

pluginloader_t

testplugin::ac_parser_t

Public Member Functions

•algo_comm_class_t ()

• virtual ∼algo_comm_class_t ()

•algo_comm_t get_c_handle ()

• virtual void local_insert_var (const char ∗,comm_var_t)

• virtual void local_remove_var (const char ∗)

• virtual void local_remove_ref (void ∗)

• virtual bool local_is_var (const char ∗)

• virtual void local_get_var (const char ∗,comm_var_t ∗)

• virtual std::string local_get_entries ()

• virtual comm_var_map_t::size_type size () const

Static Public Member Functions

• static int insert_var (void ∗, const char ∗,comm_var_t)

• static int insert_var_int (void ∗, const char ∗, int ∗)

• static int insert_var_ﬂoat (void ∗, const char ∗, ﬂoat ∗)

• static int remove_var (void ∗, const char ∗)

• static int remove_ref (void ∗, void ∗)

• static int is_var (void ∗, const char ∗)

• static int get_var (void ∗, const char ∗,comm_var_t ∗)

• static int get_var_int (void ∗, const char ∗, int ∗)

• static int get_var_ﬂoat (void ∗, const char ∗, ﬂoat ∗)

• static int get_entries (void ∗, char ∗, unsigned int)

• static const char ∗get_error (int)

604 CONTENTS

Public Attributes

• char ∗algo_comm_id_string

Private Attributes

•algo_comm_t ac

• int algo_comm_id_string_len

•comm_var_map_t vars

5.217.1 Constructor & Destructor Documentation

5.217.1.1 MHAKernel::algo_comm_class_t::algo_comm_class_t ( )

5.217.1.2 MHAKernel::algo_comm_class_t::∼algo_comm_class_t ( ) [virtual]

5.217.2 Member Function Documentation

5.217.2.1 algo_comm_t MHAKernel::algo_comm_class_t::get_c_handle ( )

5.217.2.2 int MHAKernel::algo_comm_class_t::insert_var (

void ∗handle,

const char ∗name,

comm_var_t var )[static]

5.217.2.3 int MHAKernel::algo_comm_class_t::insert_var_int (

void ∗handle,

const char ∗name,

int ∗ivar )[static]

5.217.2.4 int MHAKernel::algo_comm_class_t::insert_var_ﬂoat (

void ∗handle,

const char ∗name,

ﬂoat ∗ivar )[static]

5.217.2.5 int MHAKernel::algo_comm_class_t::remove_var (

void ∗handle,

const char ∗name )[static]

5.217.2.6 int MHAKernel::algo_comm_class_t::remove_ref (

void ∗handle,

void ∗ref )[static]

5.217 MHAKernel::algo_comm_class_t Class Reference 605

5.217.2.7 int MHAKernel::algo_comm_class_t::is_var (

void ∗handle,

const char ∗name )[static]

5.217.2.8 int MHAKernel::algo_comm_class_t::get_var (

void ∗handle,

const char ∗name,

comm_var_t ∗var )[static]

5.217.2.9 int MHAKernel::algo_comm_class_t::get_var_int (

void ∗handle,

const char ∗name,

int ∗ivar )[static]

5.217.2.10 int MHAKernel::algo_comm_class_t::get_var_ﬂoat (

void ∗handle,

const char ∗name,

ﬂoat ∗ivar )[static]

5.217.2.11 int MHAKernel::algo_comm_class_t::get_entries (

void ∗handle,

char ∗ret,

unsigned int len )[static]

5.217.2.12 const char ∗MHAKernel::algo_comm_class_t::get_error (

int e)[static]

5.217.2.13 void MHAKernel::algo_comm_class_t::local_insert_var (

const char ∗name,

comm_var_t var )[virtual]

5.217.2.14 void MHAKernel::algo_comm_class_t::local_remove_var (

const char ∗name )[virtual]

5.217.2.15 void MHAKernel::algo_comm_class_t::local_remove_ref (

void ∗addr )[virtual]

5.217.2.16 bool MHAKernel::algo_comm_class_t::local_is_var (

const char ∗name )[virtual]

5.217.2.17 void MHAKernel::algo_comm_class_t::local_get_var (

const char ∗name,

comm_var_t ∗var )[virtual]

5.217.2.18 std::string MHAKernel::algo_comm_class_t::local_get_entries ( ) [virtual]

606 CONTENTS

5.217.2.19 MHAKernel::comm_var_map_t::size_type MHAKernel::algo_comm_class_t::size ( ) const

[virtual]

5.217.3 Member Data Documentation

5.217.3.1 char∗MHAKernel::algo_comm_class_t::algo_comm_id_string

5.217.3.2 algo_comm_t MHAKernel::algo_comm_class_t::ac [private]

5.217.3.3 int MHAKernel::algo_comm_class_t::algo_comm_id_string_len [private]

5.217.3.4 comm_var_map_t MHAKernel::algo_comm_class_t::vars [private]

The documentation for this class was generated from the following ﬁles:

•mha_algo_comm.hh

•mha_algo_comm.cpp

5.218 MHAKernel::comm_var_map_t Class Reference

Inherits map<std::string, comm_var_t >.

Public Member Functions

• bool has_key (const std::string &name)

5.218.1 Member Function Documentation

5.218.1.1 bool MHAKernel::comm_var_map_t::has_key (

const std::string & name )[inline]

The documentation for this class was generated from the following ﬁle:

•mha_algo_comm.hh

5.219 MHAMultiSrc::base_t Class Reference 607

5.219 MHAMultiSrc::base_t Class Reference

Base class for source selection.

Inheritance diagram for MHAMultiSrc::base_t:

MHAMultiSrc::base_t

MHAMultiSrc::spectrum_t MHAMultiSrc::waveform_t

MHAPlugin::config_t

< MHAMultiSrc::channels_t >

Public Member Functions

•base_t (algo_comm_t iac)

• void select_source (const std::vector<std::string >&src, int in_channels)

Change the selection of input sources.

Protected Attributes

•algo_comm_t ac

Additional Inherited Members

5.219.1 Detailed Description

Base class for source selection.

See also

MHAMultiSrc::channel_t (p. 608)

MHAMultiSrc::channels_t (p. 609)

608 CONTENTS

5.219.2 Constructor & Destructor Documentation

5.219.2.1 MHAMultiSrc::base_t::base_t (

algo_comm_t iac )

5.219.3 Member Function Documentation

5.219.3.1 void MHAMultiSrc::base_t::select_source (

const std::vector<std::string >&src,

int in_channels )

Change the selection of input sources.

This function is real-time and thread safe.

Parameters

src List of input sources

in_channels Number of input channels in direct input (the processed signal)

5.219.4 Member Data Documentation

5.219.4.1 algo_comm_t MHAMultiSrc::base_t::ac [protected]

The documentation for this class was generated from the following ﬁles:

•mha_multisrc.h

•mha_multisrc.cpp

5.220 MHAMultiSrc::channel_t Class Reference

Public Attributes

• std::string name

• int channel

5.220.1 Member Data Documentation

5.220.1.1 std::string MHAMultiSrc::channel_t::name

5.220.1.2 int MHAMultiSrc::channel_t::channel

The documentation for this class was generated from the following ﬁle:

•mha_multisrc.h

5.221 MHAMultiSrc::channels_t Class Reference 609

5.221 MHAMultiSrc::channels_t Class Reference

Inherits vector<MHAMultiSrc::channel_t >.

Public Member Functions

•channels_t (const std::vector<std::string >&src, int in_channels)

Separate a list of input sources into a parsable channel list.

5.221.1 Constructor & Destructor Documentation

5.221.1.1 MHAMultiSrc::channels_t::channels_t (

const std::vector<std::string >&route,

int in_channels )

Separate a list of input sources into a parsable channel list.

The number of input channels if veriﬁed, a list of MHAMultiSrc::channel_t (p. 608) is ﬁlled.

Parameters

route vector of source channel ids

in_channels number of channels in the processed input signal

The documentation for this class was generated from the following ﬁles:

•mha_multisrc.h

•mha_multisrc.cpp

610 CONTENTS

5.222 MHAMultiSrc::spectrum_t Class Reference

Inheritance diagram for MHAMultiSrc::spectrum_t:

MHAMultiSrc::spectrum_t

MHA_AC::spectrum_t

MHASignal::spectrum_t

mha_spec_t

MHAMultiSrc::base_t

MHAPlugin::config_t

< MHAMultiSrc::channels_t >

Public Member Functions

•spectrum_t (algo_comm_t iac, std::string name, unsigned int frames, unsigned int

channels)

•mha_spec_t ∗update (mha_spec_t ∗s)

Update data of spectrum to hold actual input data.

Additional Inherited Members

5.222.1 Constructor & Destructor Documentation

5.222.1.1 MHAMultiSrc::spectrum_t::spectrum_t (

algo_comm_t iac,

std::string name,

unsigned int frames,

unsigned int channels )

5.222.2 Member Function Documentation

5.222.2.1 mha_spec_t ∗MHAMultiSrc::spectrum_t::update (

mha_spec_t ∗s)

Update data of spectrum to hold actual input data.

5.223 MHAMultiSrc::waveform_t Class Reference 611

Parameters

sInput signal chunk

Returns

Return pointer to spectrum structure

The documentation for this class was generated from the following ﬁles:

•mha_multisrc.h

•mha_multisrc.cpp

5.223 MHAMultiSrc::waveform_t Class Reference

Inheritance diagram for MHAMultiSrc::waveform_t:

MHAMultiSrc::waveform_t

MHA_AC::waveform_t

MHASignal::waveform_t

mha_wave_t

MHAMultiSrc::base_t

MHAPlugin::config_t

< MHAMultiSrc::channels_t >

Public Member Functions

•waveform_t (algo_comm_t iac, std::string name, unsigned int frames, unsigned int

channels)

•mha_wave_t ∗update (mha_wave_t ∗s)

Update data of waveform to hold actual input data.

612 CONTENTS

Additional Inherited Members

5.223.1 Constructor & Destructor Documentation

5.223.1.1 MHAMultiSrc::waveform_t::waveform_t (

algo_comm_t iac,

std::string name,

unsigned int frames,

unsigned int channels )

5.223.2 Member Function Documentation

5.223.2.1 mha_wave_t ∗MHAMultiSrc::waveform_t::update (

mha_wave_t ∗s)

Update data of waveform to hold actual input data.

Parameters

sInput signal chunk

Returns

Return pointer to waveform structure

The documentation for this class was generated from the following ﬁles:

•mha_multisrc.h

•mha_multisrc.cpp

5.224 MHAOvlFilter::band_descriptor_t Class Reference

Public Attributes

•mha_real_t cf_l

•mha_real_t ef_l

•mha_real_t cf

•mha_real_t ef_h

•mha_real_t cf_h

• bool low_side_ﬂat

• bool high_side_ﬂat

5.225 MHAOvlFilter::barkscale::bark2hz_t Class Reference 613

5.224.1 Member Data Documentation

5.224.1.1 mha_real_t MHAOvlFilter::band_descriptor_t::cf_l

5.224.1.2 mha_real_t MHAOvlFilter::band_descriptor_t::ef_l

5.224.1.3 mha_real_t MHAOvlFilter::band_descriptor_t::cf

5.224.1.4 mha_real_t MHAOvlFilter::band_descriptor_t::ef_h

5.224.1.5 mha_real_t MHAOvlFilter::band_descriptor_t::cf_h

5.224.1.6 bool MHAOvlFilter::band_descriptor_t::low_side_ﬂat

5.224.1.7 bool MHAOvlFilter::band_descriptor_t::high_side_ﬂat

The documentation for this class was generated from the following ﬁle:

•mha_fftfb.hh

5.225 MHAOvlFilter::barkscale::bark2hz_t Class Reference

Inheritance diagram for MHAOvlFilter::barkscale::bark2hz_t:

MHAOvlFilter::barkscale

::bark2hz_t

MHATableLookup::xy

_table_t

MHATableLookup::table_t

614 CONTENTS

Public Member Functions

•bark2hz_t ()

•∼bark2hz_t ()

Additional Inherited Members

5.225.1 Constructor & Destructor Documentation

5.225.1.1 MHAOvlFilter::barkscale::bark2hz_t::bark2hz_t ( )

5.225.1.2 MHAOvlFilter::barkscale::bark2hz_t::∼bark2hz_t ( )

The documentation for this class was generated from the following ﬁle:

•mha_fftfb.cpp

5.226 MHAOvlFilter::barkscale::hz2bark_t Class Reference

Inheritance diagram for MHAOvlFilter::barkscale::hz2bark_t:

MHAOvlFilter::barkscale

::hz2bark_t

MHATableLookup::xy

_table_t

MHATableLookup::table_t

Public Member Functions

•hz2bark_t ()

•∼hz2bark_t ()

5.227 MHAOvlFilter::fftfb_ac_info_t Class Reference 615

Additional Inherited Members

5.226.1 Constructor & Destructor Documentation

5.226.1.1 MHAOvlFilter::barkscale::hz2bark_t::hz2bark_t ( )

5.226.1.2 MHAOvlFilter::barkscale::hz2bark_t::∼hz2bark_t ( )

The documentation for this class was generated from the following ﬁle:

•mha_fftfb.cpp

5.227 MHAOvlFilter::fftfb_ac_info_t Class Reference

Public Member Functions

•fftfb_ac_info_t (const MHAOvlFilter::fftfb_t &fb, algo_comm_t ac, const std::string

&preﬁx)

• void insert ()

Private Attributes

•MHA_AC::waveform_t cfv

vector of nominal center frequencies / Hz

•MHA_AC::waveform_t efv

vector of edge frequencies / Hz

•MHA_AC::waveform_t bwv

vector of band-weigths (sum of squared fft-bin-weigths)/num_frames

•MHA_AC::waveform_t cLTASS

vector of LTASS correction

5.227.1 Constructor & Destructor Documentation

5.227.1.1 MHAOvlFilter::fftfb_ac_info_t::fftfb_ac_info_t (

const MHAOvlFilter::fftfb_t &fb,

algo_comm_t ac,

const std::string & preﬁx )

5.227.2 Member Function Documentation

5.227.2.1 void MHAOvlFilter::fftfb_ac_info_t::insert ( )

5.227.3 Member Data Documentation

5.227.3.1 MHA_AC::waveform_t MHAOvlFilter::fftfb_ac_info_t::cfv [private]

vector of nominal center frequencies / Hz

616 CONTENTS

5.227.3.2 MHA_AC::waveform_t MHAOvlFilter::fftfb_ac_info_t::efv [private]

vector of edge frequencies / Hz

5.227.3.3 MHA_AC::waveform_t MHAOvlFilter::fftfb_ac_info_t::bwv [private]

vector of band-weigths (sum of squared fft-bin-weigths)/num_frames

5.227.3.4 MHA_AC::waveform_t MHAOvlFilter::fftfb_ac_info_t::cLTASS [private]

vector of LTASS correction

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.228 MHAOvlFilter::fftfb_t Class Reference

FFT based overlapping ﬁlter bank.

Inheritance diagram for MHAOvlFilter::fftfb_t:

MHAOvlFilter::fftfb_t

coherence::cohflt_t

fftfbpow::fftfbpow_t

MHAOvlFilter::overlap

_save_filterbank_t

multibandcompressor

::fftfb_plug_t

MHAOvlFilter::fspacing_t

MHASignal::waveform_tmha_wave_t MHAOvlFilter::overlap

_save_filterbank_analytic_t fftfilterbank::fftfb

_plug_t

Public Member Functions

•fftfb_t (MHAOvlFilter::fftfb_vars_t &par, unsigned int nfft, mha_real_t fs)

Constructor for a FFT-based overlapping ﬁlter bank.

•∼fftfb_t ()

• void apply_gains (mha_spec_t ∗s_out, const mha_spec_t ∗s_in, const mha_wave_t

∗gains)

• void get_fbpower (mha_wave_t ∗fbpow, const mha_spec_t ∗s_in)

• void get_fbpower_db (mha_wave_t ∗fbpow, const mha_spec_t ∗s_in)

• std::vector<mha_real_t >get_ltass_gain_db () const

• unsigned int bin1 (unsigned int band) const

Return index of ﬁrst non-zero ﬁlter shape window.

• unsigned int bin2 (unsigned int band) const

Return index of ﬁrst zero ﬁlter shape window above center frequency.

• unsigned int get_fftlen () const

Return fft length.

•mha_real_t w (unsigned int k, unsigned int b) const

Return ﬁlter shape window at index k in band b.

5.228 MHAOvlFilter::fftfb_t Class Reference 617

Private Attributes

• unsigned int ∗vbin1

• unsigned int ∗vbin2

•mha_real_t(∗shape )(mha_real_t)

• unsigned int fftlen

•mha_real_t samplingrate

Additional Inherited Members

5.228.1 Detailed Description

FFT based overlapping ﬁlter bank.

5.228.2 Constructor & Destructor Documentation

5.228.2.1 MHAOvlFilter::fftfb_t::fftfb_t (

MHAOvlFilter::fftfb_vars_t &par,

unsigned int nfft,

mha_real_t fs )

Constructor for a FFT-based overlapping ﬁlter bank.

Parameters

par Parameters for the FFT ﬁlterbank that can not be deduced from the signal dimensions

are taken from this set of conﬁguration variables.

nfft FFT length

fs Sampling rate / Hz

5.228.2.2 MHAOvlFilter::fftfb_t::∼fftfb_t ( )

5.228.3 Member Function Documentation

5.228.3.1 void MHAOvlFilter::fftfb_t::apply_gains (

mha_spec_t ∗s_out,

const mha_spec_t ∗s_in,

const mha_wave_t ∗gains )

5.228.3.2 void MHAOvlFilter::fftfb_t::get_fbpower (

mha_wave_t ∗fbpow,

const mha_spec_t ∗s_in )

618 CONTENTS

5.228.3.3 void MHAOvlFilter::fftfb_t::get_fbpower_db (

mha_wave_t ∗fbpow,

const mha_spec_t ∗s_in )

5.228.3.4 std::vector<ﬂoat >MHAOvlFilter::fftfb_t::get_ltass_gain_db ( ) const

5.228.3.5 unsigned int MHAOvlFilter::fftfb_t::bin1 (

unsigned int band ) const [inline]

Return index of ﬁrst non-zero ﬁlter shape window.

5.228.3.6 unsigned int MHAOvlFilter::fftfb_t::bin2 (

unsigned int band ) const [inline]

Return index of ﬁrst zero ﬁlter shape window above center frequency.

5.228.3.7 unsigned int MHAOvlFilter::fftfb_t::get_fftlen ( ) const [inline]

Return fft length.

5.228.3.8 mha_real_t MHAOvlFilter::fftfb_t::w (

unsigned int k,

unsigned int b) const [inline]

Return ﬁlter shape window at index k in band b.

Parameters

kFrequency index

bBand index

5.228.4 Member Data Documentation

5.228.4.1 unsigned int∗MHAOvlFilter::fftfb_t::vbin1 [private]

5.228.4.2 unsigned int∗MHAOvlFilter::fftfb_t::vbin2 [private]

5.228.4.3 mha_real_t(∗MHAOvlFilter::fftfb_t::shape) (mha_real_t)[private]

5.228.4.4 unsigned int MHAOvlFilter::fftfb_t::fftlen [private]

5.228.4.5 mha_real_t MHAOvlFilter::fftfb_t::samplingrate [private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.229 MHAOvlFilter::fftfb_vars_t Class Reference 619

5.229 MHAOvlFilter::fftfb_vars_t Class Reference

Set of conﬁguration variables for FFT-based overlapping ﬁlters.

Inheritance diagram for MHAOvlFilter::fftfb_vars_t:

MHAOvlFilter::fftfb

_vars_t

coherence::vars_t

fftfbpow::fftfbpow

_interface_t

MHAOvlFilter::overlap

_save_filterbank_t::vars_t

multibandcompressor

::interface_t

fftfilterbank::fftfb

_interface_t

Public Member Functions

•fftfb_vars_t (MHAParser::parser_t &p)

construct a set of openMHA conﬁguration language variables suitable for conﬁguring the FFT-

based overlapping ﬁlterbank.

Public Attributes

•scale_var_t fscale

Frequency scale type (lin/bark/log/erb).

•scale_var_t ovltype

Filter shape (rect/lin/hann).

•MHAParser::ﬂoat_t plateau

relative plateau width.

•MHAParser::kw_t ftype

Flag to decide wether edge or center frequencies are used.

•fscale_t f

Frequency.

•MHAParser::bool_t normalize

Normalize sum of channels.

•MHAParser::bool_t fail_on_nonmonotonic

Fail if frequency entries are non-monotonic (otherwise sort)

•MHAParser::bool_t fail_on_unique_bins

Fail if center frequencies share the same FFT bin.

620 CONTENTS

•MHAParser::vﬂoat_mon_t cf

Final center frequencies in Hz.

•MHAParser::vﬂoat_mon_t ef

Final edge frequencies in Hz.

•MHAParser::vﬂoat_mon_t cLTASS

Bandwidth correction for LTASS noise (level of 0 dB RMS LTASS noise)

•MHAParser::mﬂoat_mon_t shapes

5.229.1 Detailed Description

Set of conﬁguration variables for FFT-based overlapping ﬁlters.

This class enables easy conﬁguration of the FFT-based overlapping ﬁlterbank. An instance of

fftfb_vars_t (p. 619) creates openMHA conﬁguration language variables needed for conﬁguring

the ﬁlterbank, and inserts these variables in the openMHA conﬁguration tree.

This way, the variables are visible to the user and can be conﬁgured using the openMHA con-

ﬁguration language.

5.229.2 Constructor & Destructor Documentation

5.229.2.1 MHAOvlFilter::fftfb_vars_t::fftfb_vars_t (

MHAParser::parser_t &p)

construct a set of openMHA conﬁguration language variables suitable for conﬁguring the FFT-

based overlapping ﬁlterbank.

Parameters

pThe node of the conﬁguration tree where the variables created by this instance are

inserted.

5.229.3 Member Data Documentation

5.229.3.1 scale_var_t MHAOvlFilter::fftfb_vars_t::fscale

Frequency scale type (lin/bark/log/erb).

5.229.3.2 scale_var_t MHAOvlFilter::fftfb_vars_t::ovltype

Filter shape (rect/lin/hann).

5.229 MHAOvlFilter::fftfb_vars_t Class Reference 621

5.229.3.3 MHAParser::ﬂoat_t MHAOvlFilter::fftfb_vars_t::plateau

relative plateau width.

5.229.3.4 MHAParser::kw_t MHAOvlFilter::fftfb_vars_t::ftype

Flag to decide wether edge or center frequencies are used.

5.229.3.5 fscale_t MHAOvlFilter::fftfb_vars_t::f

Frequency.

5.229.3.6 MHAParser::bool_t MHAOvlFilter::fftfb_vars_t::normalize

Normalize sum of channels.

5.229.3.7 MHAParser::bool_t MHAOvlFilter::fftfb_vars_t::fail_on_nonmonotonic

Fail if frequency entries are non-monotonic (otherwise sort)

5.229.3.8 MHAParser::bool_t MHAOvlFilter::fftfb_vars_t::fail_on_unique_bins

Fail if center frequencies share the same FFT bin.

5.229.3.9 MHAParser::vﬂoat_mon_t MHAOvlFilter::fftfb_vars_t::cf

Final center frequencies in Hz.

5.229.3.10 MHAParser::vﬂoat_mon_t MHAOvlFilter::fftfb_vars_t::ef

Final edge frequencies in Hz.

5.229.3.11 MHAParser::vﬂoat_mon_t MHAOvlFilter::fftfb_vars_t::cLTASS

Bandwidth correction for LTASS noise (level of 0 dB RMS LTASS noise)

5.229.3.12 MHAParser::mﬂoat_mon_t MHAOvlFilter::fftfb_vars_t::shapes

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

622 CONTENTS

5.230 MHAOvlFilter::fscale_bw_t Class Reference

Inheritance diagram for MHAOvlFilter::fscale_bw_t:

MHAOvlFilter::fscale_bw_t

MHAOvlFilter::fscale_t

Public Member Functions

•fscale_bw_t (MHAParser::parser_t &parent)

• std::vector<mha_real_t >get_bw_hz () const

Protected Attributes

•MHAParser::vﬂoat_t bw

•MHAParser::vﬂoat_mon_t bw_hz

Private Member Functions

• void update_hz ()

Private Attributes

•MHAEvents::connector_t<fscale_bw_t >updater

5.231 MHAOvlFilter::fscale_t Class Reference 623

Additional Inherited Members

5.230.1 Constructor & Destructor Documentation

5.230.1.1 MHAOvlFilter::fscale_bw_t::fscale_bw_t (

MHAParser::parser_t &parent )

5.230.2 Member Function Documentation

5.230.2.1 std::vector<mha_real_t >MHAOvlFilter::fscale_bw_t::get_bw_hz ( ) const

5.230.2.2 void MHAOvlFilter::fscale_bw_t::update_hz ( ) [private]

5.230.3 Member Data Documentation

5.230.3.1 MHAParser::vﬂoat_t MHAOvlFilter::fscale_bw_t::bw [protected]

5.230.3.2 MHAParser::vﬂoat_mon_t MHAOvlFilter::fscale_bw_t::bw_hz [protected]

5.230.3.3 MHAEvents::connector_t<fscale_bw_t>MHAOvlFilter::fscale_bw_t::updater

[private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.231 MHAOvlFilter::fscale_t Class Reference

Inheritance diagram for MHAOvlFilter::fscale_t:

MHAOvlFilter::fscale_t

MHAOvlFilter::fscale_bw_t

624 CONTENTS

Public Member Functions

•fscale_t (MHAParser::parser_t &parent)

• std::vector<mha_real_t >get_f_hz () const

Public Attributes

•scale_var_t unit

•MHAParser::vﬂoat_t f

•MHAParser::vﬂoat_mon_t f_hz

Private Member Functions

• void update_hz ()

Private Attributes

•MHAEvents::connector_t<fscale_t >updater

5.231.1 Constructor & Destructor Documentation

5.231.1.1 MHAOvlFilter::fscale_t::fscale_t (

MHAParser::parser_t &parent )

5.231.2 Member Function Documentation

5.231.2.1 std::vector<mha_real_t >MHAOvlFilter::fscale_t::get_f_hz ( ) const

5.231.2.2 void MHAOvlFilter::fscale_t::update_hz ( ) [private]

5.231.3 Member Data Documentation

5.231.3.1 scale_var_t MHAOvlFilter::fscale_t::unit

5.231.3.2 MHAParser::vﬂoat_t MHAOvlFilter::fscale_t::f

5.231.3.3 MHAParser::vﬂoat_mon_t MHAOvlFilter::fscale_t::f_hz

5.231.3.4 MHAEvents::connector_t<fscale_t>MHAOvlFilter::fscale_t::updater [private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.232 MHAOvlFilter::fspacing_t Class Reference 625

5.232 MHAOvlFilter::fspacing_t Class Reference

Class for frequency spacing, used by ﬁlterbank shape generator class.

Inheritance diagram for MHAOvlFilter::fspacing_t:

MHAOvlFilter::fspacing_t MHAOvlFilter::fftfb_t

coherence::cohflt_t

fftfbpow::fftfbpow_t

MHAOvlFilter::overlap

_save_filterbank_t

multibandcompressor

::fftfb_plug_t

MHAOvlFilter::overlap

_save_filterbank_analytic_t fftfilterbank::fftfb

_plug_t

Public Member Functions

•fspacing_t (const MHAOvlFilter::fftfb_vars_t &par, unsigned int nfft, mha_real_t fs)

• std::vector<unsigned int >get_cf_fftbin () const

• std::vector<mha_real_t >get_cf_hz () const

• std::vector<mha_real_t >get_ef_hz () const

• unsigned int nbands () const

Return number of bands in ﬁlter bank.

Protected Member Functions

• void fail_on_nonmonotonic_cf ()

• void fail_on_unique_fftbins ()

Protected Attributes

• std::vector<MHAOvlFilter::band_descriptor_t >bands

•mha_real_t(∗symmetry_scale )(mha_real_t)

Private Member Functions

• void ef2bands (std::vector<mha_real_t >vef)

• void cf2bands (std::vector<mha_real_t >vcf)

• void equidist2bands (std::vector<mha_real_t >vcf)

Private Attributes

• unsigned int nfft_

•mha_real_t fs_

626 CONTENTS

5.232.1 Detailed Description

Class for frequency spacing, used by ﬁlterbank shape generator class.

5.232.2 Constructor & Destructor Documentation

5.232.2.1 MHAOvlFilter::fspacing_t::fspacing_t (

const MHAOvlFilter::fftfb_vars_t &par,

unsigned int nfft,

mha_real_t fs )

5.232.3 Member Function Documentation

5.232.3.1 std::vector<unsigned int >MHAOvlFilter::fspacing_t::get_cf_fftbin ( ) const

5.232.3.2 std::vector<mha_real_t >MHAOvlFilter::fspacing_t::get_cf_hz ( ) const

5.232.3.3 std::vector<mha_real_t >MHAOvlFilter::fspacing_t::get_ef_hz ( ) const

5.232.3.4 unsigned int MHAOvlFilter::fspacing_t::nbands ( ) const [inline]

Return number of bands in ﬁlter bank.

5.232.3.5 void MHAOvlFilter::fspacing_t::fail_on_nonmonotonic_cf ( ) [protected]

5.232.3.6 void MHAOvlFilter::fspacing_t::fail_on_unique_fftbins ( ) [protected]

5.232.3.7 void MHAOvlFilter::fspacing_t::ef2bands (

std::vector<mha_real_t >vef )[private]

5.232.3.8 void MHAOvlFilter::fspacing_t::cf2bands (

std::vector<mha_real_t >vcf )[private]

5.232.3.9 void MHAOvlFilter::fspacing_t::equidist2bands (

std::vector<mha_real_t >vcf )[private]

5.232.4 Member Data Documentation

5.232.4.1 std::vector<MHAOvlFilter::band_descriptor_t>MHAOvlFilter::fspacing_t::bands

[protected]

5.232.4.2 mha_real_t(∗MHAOvlFilter::fspacing_t::symmetry_scale) (mha_real_t)[protected]

5.232.4.3 unsigned int MHAOvlFilter::fspacing_t::nfft_ [private]

5.232.4.4 mha_real_t MHAOvlFilter::fspacing_t::fs_ [private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.233 MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t Class

Reference 627

5.233 MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t Class Reference

Inheritance diagram for MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t:

MHAOvlFilter::overlap

_save_filterbank_analytic_t

fftfilterbank::fftfb

_plug_t

MHAOvlFilter::overlap

_save_filterbank_t

MHAOvlFilter::fftfb_t

MHAOvlFilter::fspacing_t MHASignal::waveform_t

mha_wave_t

MHAFilter::fftfilterbank_t

Public Member Functions

•overlap_save_ﬁlterbank_analytic_t (MHAOvlFilter::overlap_save_ﬁlterbank_t←-

::vars_t &fbpar, mhaconﬁg_t channelconﬁg_in)

• void ﬁlter_analytic (const mha_wave_t ∗sIn, mha_wave_t ∗∗ﬂtRe, mha_wave_t ∗∗ﬂt←-

Im)

Private Attributes

•MHAFilter::fftﬁlterbank_t imagfb

628 CONTENTS

Additional Inherited Members

5.233.1 Constructor & Destructor Documentation

5.233.1.1 MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t::overlap_save_ﬁlterbank_analytic_t (

MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t &fbpar,

mhaconﬁg_t channelconﬁg_in )

5.233.2 Member Function Documentation

5.233.2.1 void MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t::ﬁlter_analytic (

const mha_wave_t ∗sIn,

mha_wave_t ∗∗ ﬂtRe,

mha_wave_t ∗∗ ﬂtIm )

5.233.3 Member Data Documentation

5.233.3.1 MHAFilter::fftﬁlterbank_t MHAOvlFilter::overlap_save_ﬁlterbank_analytic_t::imagfb

[private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.234 MHAOvlFilter::overlap_save_ﬁlterbank_t Class Reference

A time-domain minimal phase ﬁlter bank with frequency shapes from MHAOvlFilter::fftfb_t

(p. 616).

5.234 MHAOvlFilter::overlap_save_ﬁlterbank_t Class Reference 629

Inheritance diagram for MHAOvlFilter::overlap_save_ﬁlterbank_t:

MHAOvlFilter::overlap

_save_filterbank_t

MHAOvlFilter::overlap

_save_filterbank_analytic_t

MHAOvlFilter::fftfb_t

MHAOvlFilter::fspacing_t MHASignal::waveform_t

mha_wave_t

MHAFilter::fftfilterbank_t

fftfilterbank::fftfb

_plug_t

Classes

• class vars_t

Public Member Functions

•overlap_save_ﬁlterbank_t (MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t &fbpar,

mhaconﬁg_t channelconﬁg_in)

•mhaconﬁg_t get_channelconﬁg () const

Private Attributes

•mhaconﬁg_t channelconﬁg_out_

Additional Inherited Members

5.234.1 Detailed Description

A time-domain minimal phase ﬁlter bank with frequency shapes from MHAOvlFilter::fftfb_t

(p. 616).

630 CONTENTS

5.234.2 Constructor & Destructor Documentation

5.234.2.1 MHAOvlFilter::overlap_save_ﬁlterbank_t::overlap_save_ﬁlterbank_t (

MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t &fbpar,

mhaconﬁg_t channelconﬁg_in )

5.234.3 Member Function Documentation

5.234.3.1 mhaconﬁg_t MHAOvlFilter::overlap_save_ﬁlterbank_t::get_channelconﬁg ( ) const

[inline]

5.234.4 Member Data Documentation

5.234.4.1 mhaconﬁg_t MHAOvlFilter::overlap_save_ﬁlterbank_t::channelconﬁg_out_ [private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

5.235 MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t Class Reference

Inheritance diagram for MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t:

MHAOvlFilter::overlap

_save_filterbank_t::vars_t

fftfilterbank::fftfb

_interface_t

MHAOvlFilter::fftfb

_vars_t

Public Member Functions

•vars_t (MHAParser::parser_t &p)

5.235 MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t Class

Reference 631

Public Attributes

•MHAParser::int_t fftlen

•MHAParser::kw_t phasemodel

•MHAParser::window_t irswnd

5.235.1 Constructor & Destructor Documentation

5.235.1.1 MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t::vars_t (

MHAParser::parser_t &p)

5.235.2 Member Data Documentation

5.235.2.1 MHAParser::int_t MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t::fftlen

5.235.2.2 MHAParser::kw_t MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t::phasemodel

5.235.2.3 MHAParser::window_t MHAOvlFilter::overlap_save_ﬁlterbank_t::vars_t::irswnd

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

632 CONTENTS

5.236 MHAOvlFilter::scale_var_t Class Reference

Inheritance diagram for MHAOvlFilter::scale_var_t:

MHAOvlFilter::scale

_var_t

MHAParser::kw_t

MHAParser::variable_t

MHAParser::monitor_t

MHAParser::base_t

Public Member Functions

•scale_var_t (const std::string &help)

• void add_fun (const std::string &name, scale_fun_t ∗fun)

• std::string get_name () const

•scale_fun_t ∗get_fun () const

•mha_real_t hz2unit (mha_real_t x) const

•mha_real_t unit2hz (mha_real_t x) const

Private Attributes

• std::vector<std::string >names

• std::vector<scale_fun_t ∗>funs

5.236 MHAOvlFilter::scale_var_t Class Reference 633

Additional Inherited Members

5.236.1 Constructor & Destructor Documentation

5.236.1.1 MHAOvlFilter::scale_var_t::scale_var_t (

const std::string & help )

5.236.2 Member Function Documentation

5.236.2.1 void MHAOvlFilter::scale_var_t::add_fun (

const std::string & name,

scale_fun_t ∗fun )

5.236.2.2 std::string MHAOvlFilter::scale_var_t::get_name ( ) const [inline]

5.236.2.3 scale_fun_t∗MHAOvlFilter::scale_var_t::get_fun ( ) const [inline]

5.236.2.4 mha_real_t MHAOvlFilter::scale_var_t::hz2unit (

mha_real_t x) const

5.236.2.5 mha_real_t MHAOvlFilter::scale_var_t::unit2hz (

mha_real_t x) const

5.236.3 Member Data Documentation

5.236.3.1 std::vector<std::string>MHAOvlFilter::scale_var_t::names [private]

5.236.3.2 std::vector<scale_fun_t∗>MHAOvlFilter::scale_var_t::funs [private]

The documentation for this class was generated from the following ﬁles:

•mha_fftfb.hh

•mha_fftfb.cpp

634 CONTENTS

5.237 MHAParser::base_t Class Reference

Base class for all parser items.

Inheritance diagram for MHAParser::base_t:

MHAParser::base_t

MHAParser::c_ifc_parser_t

MHAParser::monitor_t

MHAParser::parser_t

io_lib_t

PluginLoader::mhapluginloader_t

MHAParser::bool_mon_t

MHAParser::complex

_mon_t

MHAParser::float_mon_t

MHAParser::int_mon_t

MHAParser::mcomplex

_mon_t

MHAParser::mfloat_mon_t

MHAParser::mint_mon_t

MHAParser::string_mon_t

MHAParser::variable_t

MHAParser::vcomplex

_mon_t

MHAParser::vfloat_mon_t

MHAParser::vint_mon_t

MHAParser::vstring

_mon_t

alsa_dev_par_parser_t

AuditoryProfile::parser_t

AuditoryProfile::parser

_t::ear_t

AuditoryProfile::parser

_t::fmap_t

dc::wideband_inhib

_vars_t

DynComp::dc_afterburn

_vars_t

fw_t

io_alsa_t

io_file_t

io_parser_t

io_tcp_parser_t

MHAFilter::adapt_filter_t

MHAFilter::iir_filter_t

MHAIOJack::io_jack_t

MHAIOJackdb::io_jack_t

MHAIOPortAudio::device

_info_t

MHAIOPortAudio::io

_portaudio_t

MHAParser::mhaconfig

_mon_t

MHAParser::window_t

MHAPlugin::plugin_t

< runtime_cfg_t >

MHAPlugin_Split::split_t

MHAPlugin::plugin_t

< ac2wave_t >

MHAPlugin::plugin_t

< acConcat_wave_config >

MHAPlugin::plugin_t

< acPooling_wave_config >

MHAPlugin::plugin_t

< acSteer_config >

MHAPlugin::plugin_t

< acTransform_wave_config >

MHAPlugin::plugin_t

< adm_rtconfig_t >

MHAPlugin::plugin_t

< analysepath_t >

MHAPlugin::plugin_t

< cfg_t >

MHAPlugin::plugin_t

< cohflt_t >

Classes

• class replace_t

5.237 MHAParser::base_t Class Reference 635

Public Member Functions

•base_t (const std::string &)

Constructor for base class of all parser nodes.

•base_t (const base_t &)

• virtual ∼base_t ()

• virtual std::string parse (const std::string &)

Causes this node to process a command in the openMHA conﬁguration language.

• virtual void parse (const char ∗, char ∗, unsigned int)

This function parses a command and writes the parsing result into a C character array.

• virtual void parse (const std::vector<std::string >&, std::vector<std::string >&)

• virtual std::string op_subparse (expression_t &)

• virtual std::string op_setval (expression_t &)

• virtual std::string op_query (expression_t &)

• virtual std::string query_dump (const std::string &)

• virtual std::string query_entries (const std::string &)

• virtual std::string query_perm (const std::string &)

• virtual std::string query_range (const std::string &)

• virtual std::string query_type (const std::string &)

• virtual std::string query_val (const std::string &)

• virtual std::string query_readﬁle (const std::string &)

• virtual std::string query_saveﬁle (const std::string &)

• virtual std::string query_saveﬁle_compact (const std::string &)

• virtual std::string query_savemons (const std::string &)

• virtual std::string query_listids (const std::string &)

• std::string query_version (const std::string &)

• std::string query_id (const std::string &)

• std::string query_subst (const std::string &)

• std::string query_addsubst (const std::string &)

• std::string query_help (const std::string &)

• std::string query_cmds (const std::string &)

• void set_node_id (const std::string &)

Set the identiﬁcation string of this parser node.

• void set_help (const std::string &)

Set the help comment of a variable or parser.

• void add_parent_on_insert (parser_t ∗, std::string)

• void rm_parent_on_remove (parser_t ∗)

• const std::string & fullname () const

Return the full dot-separated path name of this parser node in the openMHA conﬁguration tree.

Public Attributes

•MHAEvents::emitter_t writeaccess

Event emitted on write access.

•MHAEvents::emitter_t valuechanged

Event emitted if the value has changed.

636 CONTENTS

•MHAEvents::emitter_t readaccess

Event emitted on read access.

•MHAEvents::emitter_t prereadaccess

Event emitted on read access, before the data ﬁeld is accessed.

Protected Member Functions

• void activate_query (const std::string &, query_t)

• void notify ()

Protected Attributes

•query_map_t queries

• bool data_is_initialized

Private Types

• typedef std::vector<replace_t >repl_list_t

Private Member Functions

• void add_replace_pair (const std::string &, const std::string &)

• std::string oplist ()

Private Attributes

• std::string help

• std::string id_str

•opact_map_t operators

•repl_list_t repl_list

• bool nested_lock

•parser_t ∗parent

• std::string thefullname

5.237.1 Detailed Description

Base class for all parser items.

The key method of the parser base class is the std::string parse(const std::string&) (p. 637)

method. Parser proxy derivatives which overwrite any of the other parse() (p. 637) methods to

be the key method must make sure that the original parse() (p. 637) method utilizes the new

key method.

5.237 MHAParser::base_t Class Reference 637

5.237.2 Member Typedef Documentation

5.237.2.1 typedef std::vector<replace_t>MHAParser::base_t::repl_list_t [private]

5.237.3 Constructor & Destructor Documentation

5.237.3.1 MHAParser::base_t::base_t (

const std::string & h)

Constructor for base class of all parser nodes.

Parameters

hHelp text describing this parser node. This help text is accessible to the conﬁguration

language through the "?help" query command.

5.237.3.2 MHAParser::base_t::base_t (

const base_t &src )

5.237.3.3 MHAParser::base_t::∼base_t ( ) [virtual]

5.237.4 Member Function Documentation

5.237.4.1 std::string MHAParser::base_t::parse (

const std::string & cs )[virtual]

Causes this node to process a command in the openMHA conﬁguration language.

Parameters

cs The command to parse

Returns

The response to the command, if successful

Exceptions

MHA_Error (p. 445)If the command cannot be executed successfully. The reason for

failure is given in the message string of the exception.

Reimplemented in PluginLoader::mhapluginloader_t (p. 883), and altplugs_t (p. 215).

638 CONTENTS

5.237.4.2 void MHAParser::base_t::parse (

const char ∗cmd,

char ∗retv,

unsigned int len )[virtual]

This function parses a command and writes the parsing result into a C character array.

This base class implementation delegates to parse(const std::string &) (p. 637).

Parameters

cmd Command to be parsed

retv Buffer for the result

len Length of buffer

Reimplemented in altplugs_t (p. 215).

5.237.4.3 void MHAParser::base_t::parse (

const std::vector<std::string >&cs,

std::vector<std::string >&retv )[virtual]

5.237.4.4 std::string MHAParser::base_t::op_subparse (

expression_t & ) [virtual]

Reimplemented in MHAParser::c_ifc_parser_t (p. 649), and MHAParser::parser_t (p. 691).

5.237.4.5 std::string MHAParser::base_t::op_setval (

expression_t & ) [virtual]

Reimplemented in MHAParser::mcomplex_t (p. 675), MHAParser::mﬂoat_t (p. 679), MH←-

AParser::vcomplex_t (p. 706), MHAParser::vﬂoat_t (p. 710), MHAParser::vint_t (p. 714),

MHAParser::complex_t (p. 655), MHAParser::ﬂoat_t (p. 661), MHAParser::int_t (p. 666),

MHAParser::bool_t (p. 647), MHAParser::vstring_t (p. 718), MHAParser::string_t (p. 699),

MHAParser::kw_t (p. 671), MHAParser::variable_t (p. 701), MHAParser::c_ifc_parser_←-

t(p. 649), and MHAParser::parser_t (p. 691).

5.237.4.6 std::string MHAParser::base_t::op_query (

expression_t & ) [virtual]

Reimplemented in MHAParser::monitor_t (p. 687), MHAParser::c_ifc_parser_t (p. 649),

and MHAParser::parser_t (p. 691).

5.237.4.7 std::string MHAParser::base_t::query_dump (

const std::string & s)[virtual]

Reimplemented in MHAParser::monitor_t (p. 687), and MHAParser::parser_t (p. 691).

5.237 MHAParser::base_t Class Reference 639

5.237.4.8 std::string MHAParser::base_t::query_entries (

const std::string & s)[virtual]

Reimplemented in MHAParser::parser_t (p. 691).

5.237.4.9 std::string MHAParser::base_t::query_perm (

const std::string & s)[virtual]

Reimplemented in MHAParser::variable_t (p. 701), and MHAParser::monitor_t (p. 687).

5.237.4.10 std::string MHAParser::base_t::query_range (

const std::string & s)[virtual]

Reimplemented in MHAParser::kw_t (p. 671), and MHAParser::range_var_t (p. 694).

5.237.4.11 std::string MHAParser::base_t::query_type (

const std::string & s)[virtual]

Reimplemented in MHAParser::mcomplex_mon_t (p. 673), MHAParser::vcomplex_mon←-

_t (p. 704), MHAParser::complex_mon_t (p. 653), MHAParser::ﬂoat_mon_t (p. 659), MH←-

AParser::mﬂoat_mon_t (p. 677), MHAParser::vﬂoat_mon_t (p. 708), MHAParser::mint←-

_mon_t (p. 686), MHAParser::vint_mon_t (p. 712), MHAParser::vstring_mon_t (p. 716),

MHAParser::string_mon_t (p. 697), MHAParser::bool_mon_t (p. 645), MHAParser::int←-

_mon_t (p. 663), MHAParser::mcomplex_t (p. 675), MHAParser::mﬂoat_t (p. 679), MH←-

AParser::vcomplex_t (p. 706), MHAParser::vﬂoat_t (p. 710), MHAParser::vint_t (p. 714),

MHAParser::complex_t (p. 655), MHAParser::ﬂoat_t (p. 661), MHAParser::int_t (p. 666),

MHAParser::bool_t (p. 647), MHAParser::vstring_t (p. 718), MHAParser::string_t (p. 699),

MHAParser::kw_t (p. 671), and MHAParser::parser_t (p. 691).

5.237.4.12 std::string MHAParser::base_t::query_val (