Digital Speech Coding for Low Bit Rate Communication Systems

by Kondoz, A. M.

• ISBN13: 9780470870082
• ISBN10: 0470870087
• Edition: 2nd
• Format: Paperback
• Copyright: 2004-10-29
• Publisher: WILEY

Summary

Building on the success of the first edition Digital Speech offers extensive new, updated and revised material based upon the latest research. This Second Edition continues to provide the fundamental technical background required for low bit rate speech coding and the hottest developments in digital speech coding techniques that are applicable to evolving communication systems. Features new chapters on Pitch Estimation and Voice-Unvoiced Classification of Speech, Harmonic Speech Coding and Multimode Speech Coding Presents a comprehensively revised chapter entitled Analysis by Synthesis LPC Coding including specific examples of popular speech coders such as CELP (Code-Excited Linear Predictive) Coding Contains an updated chapter on Efficient LPC Quantization Methods including MSVQ and anti-aliasing filtering Discusses Voice Activity Detection (VAD) methods Offers expanded coverage of speech enhancement techniques such as echo cancellation and noise suppression Written by a well-known, highly respected academic, this authoritative volume will be invaluable to practising engineers, network designers, computer scientists and advanced students in communications, electrical and electronic engineering.

Author Biography

Professor Kondoz joined the university of Surrey as a PhD. student in October 1984. From 1986 to 1988 he was employed as a research fellow in the communications group. After completing his PhD, in 1988 he was appointed as a lecturer. In 1995 he became a Reader and in 1997 Professor and Deputy Director in the Centre for Communication Systems Research (CCSR).
He has been involved in teaching of digital signal processing, telecommunications theory and source coding in both undergradute and postgraduate levels. In research he has been heading Multimedia Communication Research Group since 1990. To date, Professor Kondoz has supervised 20 successful PhD students in Speech, Video and Channel coding, Source data packetisation, Error resilient speech and video transmission and Mobile multimedia communications. His current research interests are, Low bit rate speech, image and video coding error resilient video transmission, mobile multimedia communications, robust wireless ATM, real-time terminal design and implementation for mobile communications.
Outside the University, Professor Kondoz has been a member of both the IEE and IEEE. He is a CEng and served on E5. He is on EPSRC College for signal processing and communications.

Table of Contents

Preface	xiii
Acknowledgements	xv
Introduction	1	(4)
Coding Strategies and Standards	5	(18)
Introduction	5	(1)
Speech Coding Techniques	6	(3)
Parametric Coders	7	(1)
Waveform-approximating Coders	8	(1)
Hybrid Coding of Speech	8	(1)
Algorithm Objectives and Requirements	9	(3)
Quality and Capacity	9	(1)
Coding Delay	10	(1)
Channel and Background Noise Robustness	10	(1)
Complexity and Cost	11	(1)
Tandem Connection and Transcoding	11	(1)
Voiceband Data Handling	11	(1)
Standard Speech Coders	12	(6)
ITU-T Speech Coding Standard	12	(1)
European Digital Cellular Telephony Standards	13	(1)
North American Digital Cellular Telephony Standards	14	(1)
Secure Communication Telephony	14	(1)
Satellite Telephony	15	(1)
Selection of a Speech Coder	15	(3)
Summary	18	(5)
Bibliography	18	(5)
Sampling and Quantization	23	(34)
Introduction	23	(1)
Sampling	23	(3)
Scalar Quantization	26	(13)
Quantization Error	27	(1)
Uniform Quantizer	28	(1)
Optimum Quantizer	29	(3)
Logarithmic Quantizer	32	(1)
Adaptive Quantizer	33	(3)
Differential Quantizer	36	(3)
Vector Quantization	39	(15)
Distortion Measures	42	(1)
Codebook Design	43	(1)
Codebook Types	44	(8)
Training, Testing and Codebook Robustness	52	(2)
Summary	54	(3)
Bibliography	54	(3)
Speech Signal Analysis and Modelling	57	(30)
Introduction	57	(1)
Short-Time Spectral Analysis	57	(8)
Role of Windows	58	(7)
Linear Predictive Modelling of Speech Signals	65	(12)
Source Filter Model of Speech Production	65	(2)
Solutions to LPC Analysis	67	(7)
Practical Implementation of the LPC Analysis	74	(3)
Pitch Prediction	77	(7)
Periodicity in Speech Signals	77	(1)
Pitch Predictor (Filter) Formulation	78	(6)
Summary	84	(3)
Bibliography	84	(3)
Efficient LPC Quantization Methods	87	(62)
Introduction	87	(1)
Alternative Representation of LPC	87	(3)
LPC to LSF Transformation	90	(11)
Complex Root Method	95	(1)
Real Root Method	95	(3)
Ratio Filter Method	98	(2)
Chebyshev Series Method	100	(1)
Adaptive Sequential LMS Method	100	(1)
LSF to LPC Transformation	101	(2)
Direct Expansion Method	101	(1)
LPC Synthesis Filter Method	102	(1)
Properties of LSFs	103	(2)
LSF Quantization	105	(5)
Distortion Measures	106	(1)
Spectral Distortion	106	(1)
Average Spectral Distortion and Outliers	107	(1)
MSE Weighting Techniques	107	(3)
Codebook Structures	110	(7)
Split Vector Quantization	111	(2)
Multi-Stage Vector Quantization	113	(1)
Search strategies for MSVQ	114	(2)
MSVQ Codebook Training	116	(1)
MSVQ Performance Analysis	117	(4)
Codebook Structures	117	(1)
Search Techniques	117	(2)
Perceptual Weighting Techniques	119	(2)
Inter-frame Correlation	121	(9)
LSF Prediction	122	(2)
Prediction Order	124	(1)
Prediction Factor Estimation	125	(1)
Performance Evaluation of MA Prediction	126	(2)
Joint Quantization of LSFs	128	(1)
Use of MA Prediction in Joint Quantization	129	(1)
Improved LSF Estimation Through Anti-Aliasing Filtering	130	(16)
LSF Extraction	131	(4)
Advantages of Low-pass Filtering in Moving Average Prediction	135	(11)
Summary	146	(3)
Bibliography	146	(3)
Pitch Estimation and Voiced-Unvoiced Classification of Speech	149	(50)
Introduction	149	(1)
Pitch Estimation Methods	150	(28)
Time-Domain PDAs	151	(4)
Frequency-Domain PDAs	155	(3)
Time-and Frequency-Domain PDAs	158	(8)
Pre- and Post-processing Techniques	166	(12)
Voiced-Unvoiced Classification	178	(18)
Hard-Decision Voicing	178	(11)
Soft-Decision Voicing	189	(7)
Summary	196	(3)
Bibliography	197	(2)
Analysis by Synthesis LPC Coding	199	(62)
Introduction	199	(1)
Generalized AbS Coding	200	(19)
Time-Varying Filters	202	(1)
Perceptually-based Minimization Procedure	203	(3)
Excitation Signal	206	(2)
Determination of Optimum Excitation Sequence	208	(4)
Characteristics of AbS-LPC Schemes	212	(7)
Code-Excited Linear Predictive Coding	219	(39)
LPC Prediction	221	(1)
Pitch Prediction	222	(8)
Multi-Pulse Excitation	230	(8)
Codebook Excitation	238	(14)
Joint LTP and Codebook Excitation Computation	252	(3)
CELP with Post-Filtering	255	(3)
Summary	258	(3)
Bibliography	258	(3)
Harmonic Speech Coding	261	(16)
Introduction	261	(1)
Sinusoidal Analysis and Synthesis	262	(1)
Parameter Estimation	263	(5)
Voicing Determination	264	(2)
Harmonic Amplitude Estimation	266	(2)
Common Harmonic Coders	268	(7)
Sinusoidal Transform Coding	268	(2)
Improved Multi-Band Excitation, INMARSAT-M Version	270	(1)
Split-Band Linear Predictive Coding	271	(4)
Summary	275	(2)
Bibliography	275	(2)
Multimode Speech Coding	277	(80)
Introduction	277	(3)
Design Challenges of a Hybrid Coder	280	(1)
Reliable Speech Classification	281	(1)
Phase Synchronization	281	(1)
Summary of Hybrid Coders	281	(4)
Prototype Waveform Interpolation Coder	282	(1)
Combined Harmonic and Waveform Coding at Low Bit-Rates	282	(1)
A 4 kb/s Hybrid MELP/CELP Coder	283	(1)
Limitations of Existing Hybrid Coders	284	(1)
Synchronized Waveform-Matched Phase Model	285	(13)
Extraction of the Pitch Pulse Location	286	(6)
Estimation of the Pitch Pulse Shape	292	(5)
Synthesis using Generalized Cubic Phase Interpolation	297	(1)
Hybrid Encoder	298	(13)
Synchronized Harmonic Excitation	299	(2)
Advantages and Disadvantages of SWPM	301	(3)
Offset Target Modification	304	(4)
Onset Harmonic Memory Initialization	308	(1)
White Noise Excitation	309	(2)
Speech Classification	311	(8)
Open-Loop Initial Classification	312	(3)
Closed-Loop Transition Detection	315	(3)
Plosive Detection	318	(1)
Hybrid Decoder	319	(1)
Performance Evaluation	320	(2)
Quantization Issues of Hybrid Coder Parameters	322	(9)
Introduction	322	(1)
Unvoiced Excitation Quantization	323	(1)
Harmonic Excitation Quantization	323	(8)
Quantization of ACELP Excitation at Transitions	331	(1)
Variable Bit Rate Coding	331	(5)
Transition Quantization with 4 kb/s ACELP	332	(1)
Transition Quantization with 6 kb/s ACELP	332	(1)
Transition Quantization with 8 kb/s ACELP	333	(1)
Comparison	334	(2)
Acoustic Noise and Channel Error Performance	336	(14)
Performance under Acoustic Noise	337	(8)
Performance under Channel Errors	345	(4)
Performance Improvement under Channel Errors	349	(1)
Summary	350	(7)
Bibliography	351	(6)
Voice Activity Detection	357	(22)
Introduction	357	(3)
Standard VAD Methods	360	(8)
ITU-T G.729B/G.723.1A VAD	361	(1)
ETSI GSM-FR/HR/EFR VAD	361	(1)
ETSI AMR VAD	362	(1)
TIA/EIA IS-127/733 VAD	363	(1)
Performance Comparison of VADs	364	(4)
Likelihood-Ratio-Based VAD	368	(7)
Analysis and Improvement of the Likelihood Ratio Method	370	(3)
Noise Estimation Based on SLR	373	(1)
Comparison	373	(2)
Summary	375	(4)
Bibliography	375	(4)
Speech Enhancement	379	(50)
Introduction	379	(2)
Review of STSA-based Speech Enhancement	381	(21)
Spectral Subtraction	382	(2)
Maximum-likelihood Spectral Amplitude Estimation	384	(1)
Wiener Filtering	385	(1)
MMSE Spectral Amplitude Estimation	386	(1)
Spectral Estimation Based on the Uncertainty of Speech Presence	387	(2)
Comparisons	389	(3)
Discussion	392	(10)
Noise Adaptation	402	(4)
Hard Decision-based Noise Adaptation	402	(1)
Soft Decision-based Noise Adaptation	403	(1)
Mixed Decision-based Noise Adaptation	403	(1)
Comparisons	404	(2)
Echo Cancellation	406	(17)
Digital Echo Canceller Set-up	411	(2)
Echo Cancellation Formulation	413	(2)
Improved Performance Echo Cancellation	415	(8)
Summary	423	(6)
Bibliography	426	(3)
Index	429