Voice and audio compression for wireless communications

Voice and Audio Compression for Wireless Communications......Page 3
Contents......Page 7
About the Authors......Page 23
Other Wiley and IEEE Press Books on Related Topics......Page 25
Preface and Motivation......Page 27
Acknowledgements......Page 37
I Speech Signals and Waveform Coding......Page 39
1.1 Motivation of Speech Compression......Page 41
1.2 Basic Characterisation of Speech Signals......Page 42
1.3 Classification of Speech Codecs......Page 46
1.3.1.1 Time-domain Waveform Coding......Page 47
1.3.2 Vocoders......Page 48
1.4.1 Digitisation of Speech......Page 49
1.4.2 Quantisation Characteristics......Page 51
1.4.3 Quantisation Noise and Rate-distortion Theory......Page 52
1.4.4 Non-uniform Quantisation for a known PDF: Companding......Page 54
1.4.5 PDF-independent Quantisation using Logarithmic Compression......Page 56
1.4.5.1 The µ-law Compander......Page 58
1.4.5.2 The A-law Compander......Page 59
1.4.6 Optimum Non-uniform Quantisation......Page 61
1.5 Chapter Summary......Page 66
2.1 Forward-Predictive Coding......Page 67
2.2 DPCM Codec Schematic......Page 68
2.3.1 Problem Formulation......Page 69
2.3.2 Covariance Coefficient Computation......Page 71
2.3.3 Predictor Coefficient Computation......Page 72
2.4 Adaptive One-word-memory Quantisation......Page 77
2.5 DPCM Performance......Page 78
2.6.1 Background......Page 80
2.6.2 Stochastic Model Processes......Page 82
2.7.2 Adaptive Quantiser......Page 85
2.7.3 G.721 Quantiser Scale Factor Adaptation......Page 86
2.7.4 G.721 Adaptation Speed Control......Page 88
2.7.5 G.721 Adaptive Prediction and Signal Reconstruction......Page 89
2.8 Subjective and Objective Speech Quality......Page 91
2.9.1 Motivation......Page 92
2.9.2 Embedded G.727 ADPCM Coding......Page 93
2.9.3 Performance of the Embedded G.727 ADPCM Codec......Page 94
2.10 Rate-distortionin Predictive Coding......Page 100
2.11 Chapter Summary......Page 105
II Analysis-by-Synthesis Coding......Page 107
3.1 Motivation......Page 109
3.2 Analysis-by-Synthesis Codec Structure......Page 110
3.3 The Short-term Synthesis Filter......Page 111
3.4.1 Open-loop Optimisation of LTP Parameters......Page 114
3.4.2 Closed-loop Optimisation of LTP Parameters......Page 118
3.5 Excitation Models......Page 123
3.6 Adaptive Short-term and Long-term Post-Filtering......Page 126
3.7 Lattice-based Linear Prediction......Page 128
3.8 Chapter Summary......Page 135
4.1 Log-area Ratios......Page 137
4.2.1 Derivation of the Line Spectral Frequencies......Page 141
4.2.2 Computation of the Line Spectral Frequencies......Page 145
4.2.3 Chebyshev Description of Line Spectral Frequencies......Page 147
4.3.2 Speaker-adaptive Vector Quantisation of LSFs......Page 153
4.3.3.1 Background......Page 155
4.3.3.2 The Stochastic VQ Algorithm......Page 156
4.3.4 Robust Vector Quantisation Schemes for LSFs......Page 159
4.3.5 LSF VQs in Standard Codecs......Page 160
4.4.1 Introduction to Wideband Spectral Quantisation......Page 161
4.4.1.1 Statistical Properties of Wideband LSFs......Page 163
4.4.1.2 Speech Codec Specifications......Page 165
4.4.2.1 Memoryless Vector Quantisation......Page 166
4.4.2.2 Predictive Vector Quantisation......Page 170
4.4.2.3 Multimode Vector Quantisation......Page 171
4.4.3 Simulation Results and Subjective Evaluations......Page 174
4.4.4 Conclusionson Wideband Spectral Quantisation......Page 175
4.5 Chapter Summary......Page 176
5.1 Theoretical Background......Page 177
5.2.1 Pre-processing......Page 184
5.2.3 LTP Analysis Filtering......Page 186
5.2.4 Regular Excitation Pulse Computation......Page 187
5.3 The 13 kbps RPE-LTP GSM Speech Decoder......Page 189
5.4 Bit-sensitivity of the 13 kbps GSM RPE-LTP Codec......Page 191
5.5 Application Example: A Tool-box Based Speech Transceiver......Page 192
5.6 Chapter Summary......Page 195
6.1 Background......Page 197
6.2 The Original CELP Approach......Page 198
6.3 Fixed Codebook Search......Page 201
6.4.1 Binary-pulse Excitation......Page 203
6.4.2.1 Excitation Generation......Page 204
6.4.2.2 Bit-sensitivity Analysis of the 4.8 Kbps TBPE Speech Codec......Page 206
6.4.3.1 ACELP Codebook Structure......Page 208
6.4.3.2 Dual-rate ACELP Bit Allocation......Page 210
6.4.3.3 Dual-rate ACELP Codec Performance......Page 211
6.5.1 Introduction......Page 212
6.5.2.1 Full Codebook Search Theory......Page 213
6.5.2.2 Sequential Search Procedure......Page 215
6.5.2.3 Full Search Procedure......Page 216
6.5.2.4 Sub-optimal Search Procedures......Page 218
6.5.2.5 Quantisation of the Codebook Gains......Page 219
6.5.3 Calculation of the Synthesis Filter Parameters......Page 221
6.5.3.2 Least Squares Techniques......Page 222
6.5.3.3 Optimisation via Powell’s Method......Page 225
6.5.3.4 Simulated Annealing and the Effects of Quantisation......Page 226
6.6.2.1 LSF Ordering Policies......Page 230
6.6.2.3 The Effect of Interpolation......Page 233
6.6.3 Improving the Error Sensitivity of the Excitation Parameters......Page 234
6.6.3.2 The Fixed Codebook Gain......Page 235
6.6.3.3 Adaptive Codebook Delay......Page 236
6.6.4 Matching Channel Codecs to the Speech Codec......Page 237
6.6.5 Error Resilience Conclusions......Page 241
6.7.1 The Transceiver Scheme......Page 242
6.7.2 Re-configurable Modulation......Page 243
6.7.3.1 Low-quality 3.1 kBd Mode......Page 244
6.7.3.2 High-quality 3.1 kBd Mode......Page 248
6.7.4 Voice Activity Detection and Packet Reservation Multiple Access......Page 249
6.7.5 3.1 kBd System Performance......Page 252
6.7.6 3.1 kBd System Summary......Page 255
6.8.1 Background and Motivation......Page 256
6.8.2 PRMA-assisted Multi-slot Adaptive Modulation......Page 257
6.8.3 Adaptive GSM-like Schemes......Page 258
6.8.4 Adaptive DECT-like Schemes......Page 260
6.9 Chapter Summary......Page 261
7.2.1 Introduction......Page 263
7.2.2 LPC Analysis and Quantisation......Page 265
7.2.3 The Adaptive Codebook......Page 266
7.2.4 The Fixed Codebook......Page 267
7.2.5 Error Concealment Techniques......Page 268
7.3 The 7.95 kbps Pan-American Speech Codec – Known as IS-54 DAMPS Codec......Page 269
7.4 The 6.7 kbps Japanese Digital Cellular System’s Speech Codec......Page 273
7.5.1 Introduction......Page 275
7.5.2 Codec Schematic and Bit Allocation......Page 276
7.5.3 Codec Rate Selection......Page 277
7.5.4 LPC Analysis and Quantisation......Page 278
7.5.5 The Pitch Filter......Page 279
7.5.6 The Fixed Codebook......Page 280
7.5.8 Decoder Post-filtering......Page 281
7.5.10 Conclusion......Page 282
7.6.2 Codec Schematic and Bit Allocation......Page 283
7.6.3 Encoder Pre-processing......Page 285
7.6.5 The Weighting Filter......Page 286
7.6.6 Excitation Vector 1......Page 287
7.6.7 Excitation Vector 2......Page 288
7.6.8 Channel Coding......Page 289
7.6.9 Decoder Post-processing......Page 290
7.7.1 Half-rate GSM Codec Outline and Bit Allocation......Page 291
7.7.2 Spectral Quantisation in the Half-rate GSM Codec......Page 293
7.7.3 Error Protection......Page 294
7.8.2 Codec Schematic and Bit Allocation......Page 295
7.8.3 Encoder Pre-processing......Page 296
7.8.4 LPC Analysis and Quantisation......Page 297
7.8.6 The Adaptive Codebook......Page 300
7.8.7 The Fixed Algebraic Codebook......Page 301
7.8.8 Quantisation of the Gains......Page 304
7.8.9 Decoder Post-processing......Page 305
7.8.10 G.729 Error-concealment Techniques......Page 307
7.8.11 G.729 Bit-sensitivity......Page 308
7.8.12.1 Background......Page 309
7.8.12.2 System Overview......Page 310
7.8.12.3 Turbo Channel Encoding......Page 311
7.8.12.4 OFDM in the FRAMES Speech/Data Sub-burst......Page 312
7.8.12.6 Turbo-coded G.729 OFDM Parameters......Page 313
7.8.12.7 Turbo-coded G.729 OFDM Performance......Page 314
7.8.12.8 Turbo-coded G.729 OFDM Summary......Page 315
7.9.1 Introduction......Page 316
7.9.2 The Perceptual Weighting Filter......Page 317
7.9.5 The Algebraic Codebook Search......Page 318
7.9.7 Conclusions......Page 319
7.10.1 Enhanced Full-rate GSM Codec Outline......Page 320
7.10.2.1 Spectral Quantisation and Windowing in the Enhanced Full-rate GSM Codec......Page 322
7.10.2.3 Fixed Codebook Search......Page 324
7.11.1 IS-136 Codec Outline......Page 325
7.11.2 IS-136 Bit-allocation Scheme......Page 327
7.11.3 Fixed Codebook Search......Page 328
7.11.4 IS-136 Channel Coding......Page 329
7.12.2 G.723.1 Encoding Principle......Page 330
7.12.3 Vector-quantisation of the LSPs......Page 332
7.12.4 Formant-based Weighting Filter......Page 333
7.12.5 The 6.3 kbps High-rate G.723.1 Excitation......Page 334
7.12.6 The 5.3 kbps Low-rate G.723.1 Excitation......Page 335
7.12.7 G.723.1 Bit Allocation......Page 336
7.12.8 G.723.1 Error Sensitivity......Page 338
7.13.1 Multirate Codecs and Systems......Page 340
7.13.2 System Overview......Page 343
7.13.3.1 AMR Codec Overview......Page 344
7.13.3.2 Linear Prediction Analysis......Page 345
7.13.3.5 Fixed Codebook with Algebraic Structure......Page 346
7.13.3.6 Post-processing......Page 348
7.13.3.8 Codec Mode Switching Philosophy......Page 349
7.13.4 The AMR Speech Codec’s Error Sensitivity......Page 350
7.13.5.1 RRNS Overview......Page 353
7.13.5.2 Source-matched Error Protection......Page 354
7.13.6.1 Overview......Page 356
7.13.7 System Performance......Page 357
7.13.7.1 Subjective Testing......Page 364
7.14 Chapter Summary......Page 365
8.2 Motivation and Background......Page 369
8.3 Backward-adaptive G728 Codec Schematic......Page 372
8.4.1 G728 Error Weighting......Page 374
8.4.2 G728 Windowing......Page 375
8.4.3 Codebook Gain Adaption......Page 379
8.4.4 G728 Codebook Search......Page 381
8.4.5 G728 Excitation Vector Quantisation......Page 383
8.4.6 G728 Adaptive Post-filtering......Page 385
8.4.6.1 Adaptive Long-term Post-filtering......Page 386
8.4.6.2 G.728 Adaptive Short-term Post-filtering......Page 388
8.5 Reduced-rate G728-like Codec: Variable-length Excitation Vector......Page 389
8.6 The Effects of Long-term Prediction......Page 392
8.7 Closed-loop Codebook Training......Page 397
8.8 Reduced-rate G728-like Codec: Constant-length Excitation Vector......Page 402
8.9.1 Motivation......Page 403
8.9.2 8–4 kbps Codec Improvements Due to Increasing Codebook Sizes......Page 404
8.9.3 8–4 kbps Codecs – Forward Adaption of the Short-term Synthesis Filter......Page 405
8.9.4.1 Initial Experiments......Page 406
8.9.4.2 Quantisation of Jointly Optimized Gains......Page 408
8.9.4.3 8–4 kbps Codecs – Voiced/Unvoiced Codebooks......Page 411
8.9.5 Low-delay Codecs at 4–8 kbps......Page 413
8.9.6 Low-delay ACELP Codec......Page 416
8.10.1 The Error Sensitivity of the G728 Codec......Page 419
8.10.2 The Error Sensitivity of our 4–8 kbps Low-delay Codecs......Page 420
8.10.3 The Error Sensitivity of our Low-delay ACELP Codec......Page 425
8.11.2 8–16 kbps Codec Performance......Page 426
8.11.3.1 Higher-quality Mode......Page 427
8.11.4 Speech Transceiver Performance......Page 429
8.12 Chapter Summary......Page 430
III Wideband Speech, MPEG-4 Audio and Their Transmission......Page 431
9.1.1 Introduction and Specifications......Page 433
9.1.2 G722 Codec Outline......Page 434
9.1.3 Principles of Sub-band Coding......Page 437
9.1.4.1 Analysis Filtering......Page 438
9.1.4.2 Synthesis Filtering......Page 441
9.1.4.3 Practical QMF Design Constraints......Page 443
9.1.5 G722 Adaptive Quantisation and Prediction......Page 448
9.1.6 G722 Coding Performance......Page 450
9.2.2 Transform-coding Algorithm......Page 451
9.3.1 Background......Page 454
9.3.2.2 Low-band Coding......Page 455
9.3.2.3 High-band Coding......Page 456
9.3.2.4 Bit-allocation Scheme......Page 457
9.4.1 Wideband ACELP Excitation......Page 458
9.4.2 Backward-adaptive 32 kbps Wideband ACELP......Page 460
9.4.3 Forward-adaptive 9.6 kbps Wideband ACELP......Page 461
9.5.1 Background and Motivation......Page 463
9.5.3 System Parameters......Page 466
9.5.4 Constant Throughput Adaptive Modulation......Page 467
9.5.5 Adaptive Wideband Transceiver Performance......Page 469
9.5.6 Multi-mode Transceiver Adaptation......Page 470
9.5.7 Transceiver Mode Switching......Page 471
9.5.8.1 Audio Codec Overview......Page 473
9.5.9 Detailed Description of the Audio Codec......Page 475
9.5.10 Wideband Adaptive System Performance......Page 477
9.5.11 Audio Frame Error Results......Page 478
9.5.12 Audio SEGSNR Performance and Discussions......Page 479
9.6.1 Introduction......Page 480
9.6.3 System Model......Page 483
9.6.4 Design of Irregular Convolutional Codes......Page 484
9.6.5 An Irregular Convolutional Code Example......Page 487
9.6.6 UEP AMR IRCC Performance Results......Page 488
9.6.7 UEP AMR Conclusions......Page 490
9.7.1 Introduction......Page 492
9.7.2 Audio Requirements in Mobile Multimedia Applications......Page 494
9.7.2.2 Bit Rates Supported by the Radio Network......Page 495
9.7.3 Overview of the AMR-WB+ Codec......Page 497
9.7.3.2 Stereo Encoding......Page 500
9.7.4 Performance of AMR-WB+......Page 501
9.7.5 Summary of the AMR-WB+ Codec......Page 503
9.8 Chapter Summary......Page 504
10.1 Overview of MPEG-4 Audio......Page 507
10.2 General Audio Coding......Page 509
10.2.1 Advanced Audio Coding......Page 517
10.2.3 Psycho-acoustic Model......Page 520
10.2.4 Temporal Noise Shaping......Page 522
10.2.5 Stereophonic Coding......Page 524
10.2.6 AAC Quantisation and Coding......Page 525
10.2.7 Noiseless Huffman Coding......Page 527
10.2.8 Bit-sliced Arithmetic Coding......Page 528
10.2.9 Transform-domain Weighted Interleaved Vector Quantisation......Page 530
10.3 Speech Coding in MPEG-4 Audio......Page 533
10.3.1 Harmonic Vector Excitation Coding......Page 534
10.3.2 CELP Coding in MPEG-4......Page 536
10.3.3 LPC Analysis and Quantisation......Page 538
10.3.4 Multi Pulse and Regular Pulse Excitation......Page 540
10.4 MPEG-4 Codec Performance......Page 541
10.5 MPEG-4 Space–time Block Coded OFDM Audio Transceiver......Page 543
10.5.1 System Overview......Page 544
10.5.3 Frame Dropping Procedure......Page 545
10.5.4 Space–time Coding......Page 548
10.5.5 Adaptive Modulation......Page 551
10.5.6 System Performance......Page 552
10.6.1 Motivation and Background......Page 554
10.6.2 Audio Turbo Transceiver Overview......Page 556
10.6.3 The Turbo Transceiver......Page 557
10.6.4 Turbo Transceiver Performance Results......Page 559
10.6.5 MPEG-4 Turbo Transceiver Summary......Page 562
10.7.1 Motivation and Background......Page 563
10.7.2 The AMR-WB Codec’s Error Sensitivity......Page 564
10.7.3 The MPEG-4 TWINVQ Codec’s Error Sensitivity......Page 565
10.7.4 The Turbo Transceiver......Page 566
10.7.5 Performance Results......Page 569
10.8 Chapter Summary......Page 572
IV Very Low-rate Coding and Transmission......Page 575
11.1 Low-bitrate Speech Coding......Page 577
11.1.1 AbS Coding......Page 580
11.1.2 Speech Coding at 2.4 kbps......Page 581
11.1.2.1 Background to 2.4 kbps Speech Coding......Page 582
11.1.2.2 Frequency Selective Harmonic Coder......Page 583
11.1.2.3 Sinusoidal Transform Coder......Page 584
11.1.2.4 Multiband Excitation Coders......Page 585
11.1.2.6 Mixed Excitation Linear Prediction Coder......Page 587
11.1.2.7 Waveform Interpolation Coder......Page 589
11.1.3 Speech Coding Below 2.4 kbps......Page 590
11.2 Linear Predictive Coding Model......Page 591
11.2.1 Short-term Prediction......Page 592
11.2.3 Final Analysis-by-Synthesis Model......Page 594
11.3.1 Objective Speech Quality Measures......Page 595
11.3.3 2.4 kbps Selection Process......Page 596
11.4 Speech Database......Page 598
11.5 Chapter Summary......Page 601
12.1 Overview of a Linear Predictive Vocoder......Page 603
12.2.1 Line Spectrum Frequencies Scalar Quantisation......Page 604
12.2.2 Line Spectrum Frequencies Vector Quantisation......Page 606
12.3 Pitch Detection......Page 609
12.3.1 Voiced–Unvoiced Decision......Page 611
12.3.2 Oversampled Pitch Detector......Page 612
12.3.3 Pitch Tracking......Page 616
12.3.3.1 Computational Complexity......Page 619
12.3.4 Integer Pitch Detector......Page 620
12.4 Unvoiced Frames......Page 621
12.5 Voiced Frames......Page 622
12.6 Adaptive Postfilter......Page 623
12.7.1 Pulse Dispersion Principles......Page 626
12.7.2 Pitch Independent Glottal Pulse Shaping Filter......Page 627
12.8 Results for Linear Predictive Vocoder......Page 630
12.9 Chapter Summary......Page 635
13.1.1 Fourier Theory......Page 637
13.1.3 Detecting Discontinuities with Wavelets......Page 639
13.2 Introduction to Wavelet Mathematics......Page 640
13.2.1 Multiresolution Analysis......Page 641
13.2.2 Polynomial Spline Wavelets......Page 642
13.2.3 Pyramidal Algorithm......Page 643
13.3 Preprocessing the Wavelet Transform Signal......Page 645
13.3.1 Spurious Pulses......Page 647
13.4 Voiced–unvoiced Decision......Page 648
13.5 Wavelet-based Pitch Detector......Page 650
13.5.1 Dynamic Programming......Page 651
13.5.2 Autocorrelation Simplification......Page 654
13.6 Chapter Summary......Page 657
14.1 Introduction......Page 659
14.2.1 Coding Scenarios......Page 660
14.2.1.2 U–U–V Encoder Scenario......Page 662
14.2.1.5 V–V–V Encoder Scenario......Page 663
14.2.1.8 V–V–U Encoder Scenario......Page 664
14.3 Zinc Function Modelling......Page 665
14.3.1 Error Minimisation......Page 666
14.3.2 Computational Complexity......Page 667
14.3.3 Reducing the Complexity of Zinc Function Excitation Optimisation......Page 668
14.4 Pitch Detection......Page 669
14.4.1 Voiced–unvoiced Boundaries......Page 670
14.4.2 Pitch Prototype Selection......Page 671
14.5 Voiced Speech......Page 673
14.5.1 Energy Scaling......Page 674
14.5.2 Quantisation......Page 676
14.6 Excitation Interpolation Between Prototype Segments......Page 677
14.6.1 ZFE Interpolation Regions......Page 678
14.6.3 ZFE Position Parameter Interpolation......Page 680
14.6.5 Removal of ZFE Pulse Position Signalling and Interpolation......Page 682
14.8 Adaptive Postfilter......Page 683
14.9 Results for Single Zinc Function Excitation......Page 684
14.10 Error Sensitivity of the 1.9 kbps PWI-ZFE Coder......Page 687
14.10.1.2 Voiced–unvoiced Flag......Page 688
14.10.1.6 Boundary Shift Parameter......Page 689
14.10.2 Degradation from Bit Corruption......Page 690
14.10.2.1 Error Sensitivity Classes......Page 691
14.11.1 Encoding Algorithm......Page 692
14.11.2 Performance of Multiple Zinc Function Excitation......Page 695
14.12.1 Motivation......Page 699
14.12.3 Turbo Channel Coding......Page 700
14.12.4 The Turbo-coded GMSK Transceiver......Page 702
14.13 Chapter Summary......Page 703
15.1 Introduction......Page 705
15.2 Overview of Mixed-multiband Excitation......Page 706
15.3 Finite Impulse Response Filter......Page 709
15.4 Mixed-multiband Excitation Encoder......Page 711
15.4.1 Voicing Strengths......Page 712
15.5 Mixed-multiband Excitation Decoder......Page 714
15.5.1 Adaptive Postfilter......Page 716
15.5.2 Computational Complexity......Page 717
15.6.1 Performance of a Mixed-multiband Excitation Linear Predictive Coder......Page 718
15.6.2 Performance of a Mixed-multiband Excitation and Zinc Function Prototype Excitation Coder......Page 721
15.7 A Higher Rate 3.85 kbps Mixed-multiband Excitation Scheme......Page 724
15.8.1 Background......Page 729
15.8.2 The Speech Codec’s Bit Allocation......Page 730
15.8.3 The Speech Codec’s Error Sensitivity......Page 731
15.8.4 Channel Coding......Page 732
15.8.5 The JD-CDMA Speech System......Page 733
15.8.6 System Performance......Page 734
15.9 Chapter Summary......Page 737
16.1 Introduction......Page 739
16.2.1 Sinusoidal Analysis with Peak-picking......Page 740
16.2.2 Sinusoidal Analysis using Analysis-by-synthesis......Page 741
16.3.1 Frequency, Amplitude and Phase Interpolation......Page 742
16.4 Low-bitrate Sinusoidal Coders......Page 743
16.4.2 Incorporating Linear Prediction Analysis......Page 746
16.5 Incorporating Prototype Waveform Interpolation......Page 747
16.6 Encoding the Sinusoidal Frequency Component......Page 748
16.7.1 Peak-picking of the Residual Spectra......Page 750
16.7.2 Analysis-by-synthesis of the Residual Spectrum......Page 751
16.7.4 Reducing the Computational Complexity......Page 753
16.8.1.2 Interpolation and Decimation......Page 758
16.8.1.3 Vector Quantisation......Page 759
16.8.1.4 Vector Quantisation Performance......Page 761
16.8.1.5 Scalar Quantisation of the Amplitudes......Page 762
16.8.2.2 Encoding the Phases with a Voiced–unvoiced Switch......Page 763
16.8.3.1 Equivalent Rectangular Bandwidth Scale......Page 764
16.8.4 Voiced–unvoiced Flag......Page 765
16.9 Sinusoidal Transform Decoder......Page 766
16.9.3 Computational Complexity......Page 767
16.10 Speech Coder Performance......Page 768
16.11 Chapter Summary......Page 774
17.1 Summary......Page 775
17.2 Listening Tests......Page 776
17.3 Summary of Very-low-rate Coding......Page 777
17.4 Further Research......Page 779
18.1 Background to Speech Quality Evaluation......Page 781
18.2.1 Introduction......Page 782
18.2.3 Articulation Index......Page 783
18.2.4 Cepstral Distance......Page 784
18.2.5 Example: Computation of Cepstral Coefficients......Page 788
18.2.6 Logarithmic Likelihood Ratio......Page 789
18.3.1 Quality Tests......Page 790
18.4.1 Background......Page 791
18.5 Subjective Speech Quality of Various Codecs......Page 793
18.7 Objective Speech Performance of Various Transceivers......Page 795
18.8 Chapter Summary......Page 802
19.1 Introduction......Page 803
19.2.2.1 Registration......Page 804
19.2.2.2 Call Setup......Page 806
19.2.2.3 Terminate a Call......Page 808
19.2.2.4 Cancel a Call......Page 809
19.2.3 Session Description Protocol......Page 810
19.3.1 Introduction......Page 812
19.3.2.2 Call Establishment......Page 813
19.3.2.5 Call Termination......Page 815
19.4 Real-time Transport Protocol......Page 816
19.4.2.2 RTP Payload for G.729......Page 817
19.5 Conclusion......Page 819
A Constructing the Quadratic Spline Wavelets......Page 821
B Zinc Function Excitation......Page 825
C Probability Density Function for Amplitudes......Page 831
Bibliography......Page 835
Index......Page 863
Author Index......Page 873