Speech and Audio Processing in Adverse Environments (eBook)

Preface 6
Contents 7
List of Contributors 18
Abbreviations and Acronyms 20
1 Introduction 26
1.1 Overview about the Book 27
Part I Speech Enhancement 30
2 Low Delay Filter-Banks for Speech and Audio Processing 31
2.1 Introduction 31
2.2 Analysis-Synthesis Filter-Banks 33
2.3 The Filter-Bank Equalizer 47
2.4 Further Measures for Signal Delay Reduction 62
2.5 Application to Noise Reduction 67
2.6 Conclusions 73
References 74
3 A Pre-Filter for Hands-Free Car Phone Noise Reduction: Suppression of Harmonic Engine Noise Components 80
3.1 Introduction 80
3.2 Analysis of the Different Car Noise Components 81
3.3 Engine Noise Removal Based on Notch Filters 85
3.4 Compensation of Engine Harmonics with Adaptive Filters 90
3.5 Evaluation and Comparison of the Results Obtained by the Notch Filter and the Compensation Approach 101
3.6 Conclusions and Summary 102
References 104
4 Model-Based Speech Enhancement 105
4.1 Introduction 105
4.2 Conventional Speech Enhancement Schemes 107
4.3 Speech Enhancement Schemes Based on Nonlinearities 109
4.4 Speech Enhancement Schemes Based on Speech Reconstruction 113
4.5 Combining the Reconstructed and the Noise Suppressed Signal 140
4.6 Summary and Outlook 149
References 149
5 Bandwidth Extension of Telephony Speech 151
5.1 Introduction 151
5.2 Organization of the Chapter 153
5.3 Basics 154
5.4 Non-Model-Based Algorithms for Bandwidth Extension 165
5.5 Model-Based Algorithms for Bandwidth Extension 169
5.6 Evaluation of Bandwidth Extension Algorithms 192
5.7 Conclusions 197
References 198
6 Dereverberation and Residual Echo Suppression in Noisy Environments 201
6.1 Introduction 202
6.2 Problem Formulation 204
6.3 OM-LSA Estimator for Multiple Interferences 207
6.4 Dereverberation of Noisy Speech Signals 211
6.5 Residual Echo Suppression 219
6.6 Joint Suppression of Reverberation, Residual Echo, and Noise 226
6.7 Experimental Results 228
6.8 Summary and Outlook 239
References 240
7 Low Distortion Noise Cancellers – Revival of a Classical Technique 244
7.1 Introduction 244
7.2 Distortions in Widrow’s Adaptive Noise Canceller 245
7.3 Paired Filter (PF) Structure 248
7.4 Crosstalk Resistant ANC and Cross-Coupled Structure 254
7.5 Cross-Coupled Paired Filter (CCPF) Structure 257
7.6 Generalized Cross-Coupled Paired Filter (GCCPF) Structure 262
7.7 Demonstration in a Personal Robot 276
7.8 Conclusions 276
References 278
Part II Echo Cancellation 280
8 Nonlinear Echo Cancellation Based on Spectral Shaping 281
8.1 Introduction 281
8.2 Frequency-Domain Model of Highly Nonlinear Residual Echo 282
8.3 Echo Canceller Based on the New Residual Echo Model 288
8.4 Evaluations 291
8.5 DSP Implementation and Real-Time Evaluation 294
8.6 Conclusions 294
References 295
Part III Signal and System Quality Evaluation 298
9 Telephone-Speech Quality 299
9.1 Telephone-Speech Signals 299
9.2 Speech-Signal Quality 301
9.3 Speech-Quality Assessment 304
9.4 Compound-System Quality Prediction 305
9.5 Auditory Total-Quality Assessment 306
9.6 Auditory Quality-Attribute Analysis 310
9.7 Instrumental Total-Quality Measurement 318
9.8 Instrumental Attribute-Based Quality Measurements 332
9.9 Conclusions, Outlook, and Final Remarks 343
References 344
10 Evaluation of Hands-free Terminals 350
10.1 Introduction 350
10.2 Quality Assessment of Hands-free Terminals 351
10.3 Subjective Methods for Determining the Communicational Quality 353
10.4 Test Environment 361
10.5 Test Signals and Analysis Methods 365
10.6 Result Representation 376
10.7 Related Aspects 379
References 386
Part IV Multi-Channel Processing 389
11 Correlation-Based TDOA-Estimation for Multiple Sources in Reverberant Environments 390
11.1 Introduction 390
11.2 Analysis of TDOA Ambiguities 392
11.3 Estimation of Direct Path TDOAs 399
11.4 Consistent TDOA Graphs 406
11.5 Experimental Results 415
11.6 Summary 423
References 424
12 Microphone Calibration for Multi-Channel Signal Processing 426
12.1 Introduction 426
12.2 Beamforming with Ideal Microphones 427
12.3 Microphone Mismatch and its E.ect on Beamforming 436
12.4 Calibration Techniques and their Limits for Real- World Applications 441
12.5 Self-Calibration Techniques 458
12.6 Summary 468
12.A Experimental Determination of the Directivity Index 469
References 474
13 Convolutive Blind Source Separation for Noisy Mixtures 477
13.1 Introduction 477
13.2 Blind Source Separation for Acoustic Mixtures Based on the TRINICON Framework 481
13.3 Extensions for Blind Source Separation in Noisy Environments 498
13.4 Conclusions 526
References 527
14 Binaural Speech Segregation 533
14.1 Introduction 533
14.2 T–F Masks for CASA 536
14.3 Anechoic Binaural Segregation 537
14.4 Reverberant Binaural Segregation 541
14.5 Evaluation 544
14.6 Concluding Remarks 551
14.7 Acknowledgments 554
References 554
15 Spatio-Temporal Adaptive Inverse Filtering in the Wave Domain 558
15.1 Introduction 558
15.2 Problem Description 560
15.3 Computation of Compensation Filters 570
15.4 Eigenspace Adaptive Filtering 575
15.5 Wave-Domain Adaptive Filtering 577
15.6 Application of WDAF to Adaptive Inverse Filtering Problems 583
15.7 Conclusions 585
References 587
Part V Selected Applications 591
16 Virtual Hearing 592
16.1 Previous Work 593
16.2 VirtualHearing 597
16.3 Room Acoustic Model 598
16.4 HRTF Simulation 602
16.5 Neural Model 605
16.6 The Software and Interface 610
16.7 Software Testing 614
16.8 Future Work and Conclusions 615
References 618
17 Dynamic Sound Control Algorithms in Automobiles 620
17.1 Introduction 620
17.2 Previous Systems – Description and Analysis 624
17.3 Spectrum-Based Dynamic Equalization Control 650
17.4 Conclusion and Outlook 675
17.5 Acknowledgement 678
References 679
18 Towards Robust Distant-Talking Automatic Speech Recognition in Reverberant Environments 684
18.1 Introduction 684
18.2 The Distant-Talking ASR Scenario 685
18.3 How to Deal with Reverberation in ASR Systems? 688
18.4 Effect of Reverberation in the Feature Domain 696
18.5 Signal Dereverberation and Beamforming 700
18.6 Robust Features 704
18.7 Model Training and Adaptation 705
18.8 Reverberation Modeling for Speech Recognition 707
18.9 Summary and Conclusions 727
References 728
Index 734