Details

<p>List of Contributors xxi</p> <p>Preface xxvii</p> <p><b>1 Introduction 1</b><br /><i>Rainer Martin, Ulrich Heute, Christiane Antweiler</i></p> <p>I Speech Quality Assessment 7</p> <p><b>2 Speech-Transmission Quality: Aspects and Assessment for Wideband vs. Narrowband Signals 9</b><br /><i>Ulrich Heute</i></p> <p>2.1 Introduction 9</p> <p>2.2 Speech Signals . 10</p> <p>2.3 Telephone-Band Speech Signals 11</p> <p>2.4 Wideband Speech Signals 14</p> <p>2.5 Speech-Quality Assessment 25</p> <p>2.6 Wideband Speech-Quality Assessment 30</p> <p>2.7 Concluding Remarks 43</p> <p>Bibliography 44</p> <p><b>3 Parametric Quality Assessment of Narrowband Speech in Mobile Communication Systems 51</b><br /><i>Marc Werner</i></p> <p>3.1 Introduction 51</p> <p>3.2 Simulations of GSM and UMTS Speech Transmissions 58</p> <p>3.3 Speech Quality Measures based on Transmission Parameters 65</p> <p>3.4 Discussion and Conclusions 73</p> <p>Bibliography 73</p> <p>II Adaptive Algorithms in Acoustic Signal Processing 77</p> <p><b>4 Kalman Filtering in Acoustic Echo Control: A Smooth Ride on a Rocky Road 79</b><br /><i>Gerald Enzner</i></p> <p>4.1 Introduction 79</p> <p>4.2 A Comprehensive Theory of Acoustic Echo Control 85</p> <p>4.3 The Kalman Filter for Conditional Mean and Covariance Estimation 90</p> <p>4.4 AEC Performance of the Frequency-Domain Adaptive Kalman Filter 100</p> <p>4.5 Discussion and Conclusions 102</p> <p>Bibliography 103</p> <p><b>5 Noise Reduction - Statistical Analysis and Control of Musical Noise 107</b><br /><i>Colin Breithaupt, Rainer Martin</i></p> <p>5.1 Introduction 107</p> <p>5.2 Speech Enhancement in the DFT Domain 109</p> <p>5.3 Measurement and Assessment of Unnatural Fluctuations 115</p> <p>5.4 Avoidance of Processing Artifacts 120</p> <p>5.5 Control of Spectral Fluctuations in the Cepstral Domain 123</p> <p>5.6 Discussion and Conclusions 128</p> <p>5.7 Acknowledgements 129</p> <p>5.8 Appendix 129</p> <p>Bibliography 131</p> <p><b>6 Acoustic Source Localization with Microphone Arrays 135</b><br /><i>Nilesh Madhu, Rainer Martin</i></p> <p>6.1 Introduction 135</p> <p>6.2 SignalModel 136</p> <p>6.3 Localization Approach Taxonomy 141</p> <p>6.4 Indirect Localization Approaches 141</p> <p>6.5 Direct Localization Approaches 148</p> <p>6.6 Evaluation of Localization Algorithms 156</p> <p>6.7 Conclusions 166</p> <p>Bibliography 166<br /><br /><b>7 Multi-Channel System Identification with Perfect Sequences – Theory and Applications – 171</b><br /><i>Christiane Antweiler</i></p> <p>7.1 Introduction 171</p> <p>7.2 System Identification with Perfect Sequences 174</p> <p>7.3 Multi-Channel System Identification 185</p> <p>7.4 Applications 191</p> <p>7.5 Discussion and Conclusions 195</p> <p>Bibliography 195</p> <p>III Speech Coding for Heterogeneous Networks 199</p> <p><b>8 Embedded Speech Coding: From G.711 to G.729.1 201</b><br /><i>Bernd Geiser, Stéphane Ragot, Hervé Taddei</i></p> <p>8.1 Introduction 201</p> <p>8.2 Theory and Tools of Embedded Speech Coding 203</p> <p>8.3 Embedded Speech Coding Methods 212</p> <p>8.4 Standardized Embedded Speech Coders 219</p> <p>8.5 Network Aspects of Embedded Speech Coding 232</p> <p>8.6 Conclusions and Perspectives 237</p> <p>Bibliography 238</p> <p><b>9 Backwards Compatible Wideband Telephony 249</b><br /><i>Peter Jax</i></p> <p>9.1 Introduction 249</p> <p>9.2 From Narrowband Telephony to Wideband Telephony 250</p> <p>9.3 Stand-Alone Bandwidth Extension 254</p> <p>9.4 Embedded Wideband Coding Using Bandwidth Extension Techniques 257</p> <p>9.5 Combination of Bandwidth Extension with Watermarking 262</p> <p>9.6 Advanced Transmission of Highband Parameters 267</p> <p>9.7 Conclusions 274</p> <p>Bibliography 274</p> <p>IV Joint Source-Channel Coding 279</p> <p><b>10 Parameter Models and Estimators in Soft Decision Source Decoding 281</b><br /><i>Tim Fingscheidt</i></p> <p>10.1 Introduction 281</p> <p>10.2 Overview to Soft Decision Source Decoding 283</p> <p>10.3 The Markovian Parameter Model 287</p> <p>10.4 Basic Extrapolative Estimators 290</p> <p>10.5 Joint Extrapolative Estimation of Two Different Parameters 298</p> <p>10.6 Extrapolative Estimation with Repeated Parameter Transmission 301</p> <p>10.7 Interpolative Estimation of a Parameter 304</p> <p>10.8 Discussion and Conclusions 307</p> <p>Bibliography 307</p> <p><b>11 Optimal MMSE Estimation for Vector Sources with Spatially and Temporally Correlated Elements 311</b><br /><i>Stefan Heinen, Marc Adrat</i></p> <p>11.1 Introduction 311</p> <p>11.2 Source Model 312</p> <p>11.3 Transmission Channel 316</p> <p>11.4 Optimal MMSE Parameter Estimator 316</p> <p>11.5 Near-Optimal MMSE Parameter Estimator 320</p> <p>11.6 Illustrative Comparison 323</p> <p>11.7 Simulation Results 325</p> <p>11.8 Conclusions 327</p> <p>Bibliography 327</p> <p><b>12 Source Optimized Channel Codes & Source Controlled Channel Decoding 329</b><br /><i>Stefan Heinen, Thomas Hindelang</i></p> <p>12.1 Introduction 329</p> <p>12.2 The Transmission System Used as Reference 330</p> <p>12.3 Source Optimized Channel Coding (SOCC) 332</p> <p>12.4 Source Controlled Channel Decoding (SCCD) 341</p> <p>12.5 Comparison of SOCC versus SCCD 357</p> <p>12.6 Conclusions 362</p> <p>Bibliography 363</p> <p><b>13 Iterative Source-Channel Decoding & Turbo DeCodulation 365</b><br /><i>Marc Adrat, Thorsten Clevorn, Laurent Schmalen</i></p> <p>13.1 Introduction 365</p> <p>13.2 The Key of the Turbo Principle: Extrinsic Information 366</p> <p>13.3 Iterative Source-Channel Decoding (ISCD) 379</p> <p>13.4 Turbo DeCodulation (TDeC) 387</p> <p>13.5 Conclusions 394</p> <p>Bibliography 395</p> <p>V Speech Processing in Hearing Instruments 399</p> <p><b>14 Binaural Signal Processing in Hearing Aids 401</b><br /><i>Volkmar Hamacher, Ulrich Kornagel, Thomas Lotter, Henning Puder</i></p> <p>14.1 Introduction 401</p> <p>14.2 Wireless System for Hearing Aids 405</p> <p>14.3 Binaural Classification Systems 410</p> <p>14.4 Binaural Beamformer 415</p> <p>14.5 Blind Source Separation (BSS): An Application for a Binaural Directional Microphone Array in Hearing Aids 422</p> <p>14.6 Conclusions 427</p> <p>Bibliography 428</p> <p><b>15 Auditory-profile-based Physical Evaluation of Multi-microphone Noise Reduction Techniques in Hearing Instruments 431</b><br /><i>Koen Eneman, Arne Leijon, Simon Doclo, Ann Spriet, Marc Moonen, Jan Wouters</i></p> <p>15.1 Introduction 431</p> <p>15.2 Multi-microphone Noise Reduction in Hearing Instruments 434</p> <p>15.3 Auditory-profile-based Physical Evaluation 441</p> <p>15.4 Test Conditions 449</p> <p>15.5 Simulation Results 450</p> <p>15.6 Discussion 452</p> <p>15.7 Conclusions 455</p> <p>Bibliography 456</p> <p>VI Speech Processing for Human–Machine Interfaces 459</p> <p><b>16 Automatic Speech Recognition in Adverse Acoustic Conditions 461</b><br /><i>Hans-Günter Hirsch</i></p> <p>16.1 Introduction 461</p> <p>16.2 Structure of Speech Recognition Systems 462</p> <p>16.3 Acoustic Scenarios during Speech Input 468</p> <p>16.4 Improving the Recognition Performance in Adverse Conditions 476</p> <p>16.5 Conclusions 493</p> <p>Bibliography 494</p> <p><b>17 Speaker Classification for Next-Generation Voice-Dialog Systems 497</b><br /><i>Felix Burkhardt, Florian Metze, Joachim Stegmann</i></p> <p>17.1 Introduction 497</p> <p>17.2 Speaker Classification 498</p> <p>17.3 Detection of Age and Gender 505</p> <p>17.4 Detection of Anger 510</p> <p>17.5 Applications in IVR Systems 517</p> <p>17.6 Discussion and Conclusion 523</p> <p>Bibliography 525</p> <p>Index 529</p> <p>Permissions List 541</p>

"This book is recommended to anyone interested in the details of the latest developments in the area of digital-speech enhancement and processing." (<i>Computing Reviews</i>, August 8, 2008)

<b>Dr. Ing.</b> <b>Rainer Martin</b> is a Professor of Information Technology at Ruhr University and Head of the Institute of Communication Acoustics, Bochum, Germany. His research interests are signal processing for voice communication systems, acoustics, and human-machine interfaces. He has worked on algorithms for noise reduction, acoustic echo cancellation, microphone arrays, and speech recognition. He is coauthor of the book Vary/Martin “Digital Speech Transmission”, John Wiley, 2006. <p><b>Ulrich Heute</b> is a Professor for circuit and system theory at Christian-Albrechts University, Kiel. His research interests include digital signal processing, filters and filter banks, and spectral analysis, with applications in medical, audio, and, especially, speech-signal processing (combined source and channel coding, enhancement, modeling, speaker characterization, and instrumental quality assessment).</p> <p><b>Christiane Antweiler</b> is a senior scientist at the Institute of Communication Systems and Data Processing of the RWTH Aachen University. Her interests are the design and implementation of digital signal and speech processing algorithms for real-time applications, and her special focus lies on speech coding for cellular mobile radio and the enhancement of digital speech signals. Furthermore she is interested in algorithms for system identification, in adaptive filter theory and in digital signal processing algorithms for medical diagnostics.</p>

Speech processing and speech transmission technology are expanding fields of active research. New challenges arise from the 'anywhere, anytime' paradigm of mobile communications, the ubiquitous use of voice communication systems in noisy environments and the convergence of communication networks toward Internet based transmission protocols (such as Voice over IP). As a consequence, new speech coding, new enhancement and error concealment, and new quality assessment methods are emerging. <p><i>Advances in Digital Speech Transmission</i> provides an up-to-date overview of the field, including topics such as speech coding in heterogeneous communication networks, wideband coding, and the quality assessment of modern speech transmission systems.</p> <p><b>Provides an insight into the latest developments in speech processing and speech transmission, making it an essential reference to those working in these fields</b></p> <p><b>Offers a balanced overview of technology and applications</b></p> <p><b>Presents a unique in-depth discussion of soft-decision source decoding, joint source-channel coding and Turbo DeCodulation techniques</b></p> <p><b>Explains speech signal processing in hearing instruments and human-machine interfaces from applications point of view</b></p> <p><b>Covers speech coding for Voice over IP, source localization, digital hearing aids and speech processing for automatic speech recognition</b></p> <p><i>Advances in Digital Speech Transmission</i> serves as an essential link between the basics and the type of technology and applications (prospective) engineers work on in industry labs and academia. The book will also be of interest to advanced students, researchers, and other professionals who need to brush up their knowledge in this field.</p>

GB