Jacob Benesty, M. Mohan Sondhi, Yiteng Huang (eds.)
This handbook plays a fundamental role in sustainable progress in speech research and development. With an accessible format and with accompanying DVD-Rom, it targets three categories of readers: graduate students, professors and active researchers in academia, and engineers in industry who need to understand or implement some specific algorithms for their speech-related products. It is a superb source of application-oriented, authoritative and comprehensive information about these technologies, this work combines the established knowledge derived from research in such fast evolving disciplines as Signal Processing and Communications, Acoustics, Computer Science and Linguistics.
Foreword by J. L. Flanagan Chap. 1 Introduction to Speech Processing Part A: Production, Perception, and Modeling of Speech (M. M. Sondhi) Part A describes the contemporary views on phonatory and articulatory mechanisms of humans to illustrate the physiological processes of speech production. It also describes the nonlinear cochlear speech processing in auditory masking, the perception of speech and sound by humans, and various methods for speech quality assessment with a focus on standardized methods. Chap. 2 Physiological Processes of Speech Production Chap. 3 Nonlinear Cochlear Signal Processing and Masking in Speech Perception Chap. 4 Perception of Speech and Sound Chap. 5 Speech Quality Estimation Part B: Signal Processing for Speech (Y. Huang, J. Benesty) Part B gives a large number of signal processing concepts and algorithms that are widely used in speech processing and in the applications of speech. Chap. 6 Wiener and Adaptive Filters Chap. 7 Linear Prediction Chap. 8 Kalman Filter Chap. 9 Homomorphic Systems and Cepstrum Analysis of Speech Chap. 10 Pitch and Voicing Determination of Speech with an Extension Toward Music Signals Chap. 11 Formant Estimation and Tracking Chap. 12 The STFT, Sinusoidal Models, and Speech Modification Chap. 13 Adaptive Blind Multichannel Identification Part C: Speech Coding (W. B. Kleijn) Part C discusses the attributes of speech coders as well as the underlying principles that determine their behavior and architecture. Coders for both traditional and packet networks are discussed, as well as low-bit-rate speech coding, various speech coding standards, and perceptual audio coders. Chap. 14 Principles of Speech Coding Chap. 15 Voice over IP: Speech Transmission over Packet Networks Chap. 16 Low-Bit-Rate Speech Coding Chap. 17 Analysis-by-Synthesis Speech Coding Chap. 18 Perceptual Audio Coding of Speech Signals Part D: Text-to-Speech Synthesis (S. Narayanan) Part D presents different techniques for speech synthesis, including rule-based, corpus-based, and a combination of both. Linguistic analysis and prosodic processing, which are important parts of a text-to-speech (TTS) system, are reviewed. Other aspects of interest for TTS such as voice transformation and synthesis of expressive speech are also discussed. Chap. 19 Basic Principles of Speech Synthesis Chap. 20 Rule-Based Speech Synthesis Chap. 21 Corpus-Based Speech Synthesis Chap. 22 Linguistic Processing for Speech Synthesis Chap. 23 Prosodic Processing Chap. 24 Voice Transformation Chap. 25 Expressive/Affective Speech Synthesis Part E: Speech Recognition (L. Rabiner, B.-H. Juang) Part E describes the most important speech recognition technologies. The approach based on the powerful hidden Markov models is generously presented and some other promising approaches are outlined. The robustness issues concerning the acoustical environment are studied. Finally, several fundamental applications are also discussed. Chap. 26 Historical Perspective of the Field of ASR/NLU Chap. 27 HMMs and Related Speech Technologies Chap. 28 Speech Recognition with Weighted Finite-State Transducers Chap. 29 A Machine Learning Framework for Spoken-Dialog Classification Chap. 30 Towards Superhuman Speech Recognition Chap. 31 Natural Language Understanding Chap. 32 Transcription and Distillation of Spontaneous Speech Chap. 33 Environmental Robustness Chap. 34 The Business of Speech Technologies Chap. 35 Spoken Dialog Systems Part F: Speaker Recognition (S. Parthasarathy) Part F develops the field of speaker recognition. It covers text-dependent and text-independent speaker recognition and their applications. Chap. 36 Overview of Speaker Recognition Chap. 37 Text-Dependent Speaker Recognition Chap. 38 Text-Independent Speaker Recognition Part G: Language Recognition (C.-H. Lee) Part G provides an overview on principles of state-of-the-art language recognition approaches. Language characterization, identification, and modeling are addressed. Vector space characterization approaches to converting speech utterances into spoken document vectors for modeling and classification are also presented. Chap. 39 Principle of Spoken Language Recognition Chap. 40 Spoken Language Characterization Chap. 41 Automatic Language Recognition via Spectral and Token Based Approaches Chap. 42 Vector Based Spoken Language Classification Part H: Speech Enhancement (J. Chen, S. Gannot, J. Benesty) Part H develops all classical aspects of speech enhancement: noise reduction, dereverberation, echo cancellation, feedback control, and active noise control. Chap. 43 Fundamentals of Noise Reduction Chap. 44 Spectral Enhancement methods Chap. 45 Echo Cancellation Chap. 46 Dereverberation Chap. 47 Adaptive Beamforming and Postfiltering Chap. 48 Feedback Control in Hearing Aids Chap. 49 Active Noise Control Part I: Multichannel Speech Processing (J. Benesty, I. Cohen, Y. Huang) Part I presents modern aspects of multichannel processing, for acoustic scene analysis, speech acquisition and presentation, when a large number of microphones and loudspeakers are available. Chap. 50 Microphone Arrays Chap. 51 Time Delay Estimation and Source Localization Chap. 52 Convolutive Blind Source Separation Methods Chap. 53 Sound Field Reproduction About the Authors Subject Index