Signal processing of speech

Author(s)

Bibliographic Information

Signal processing of speech

F.J. Owens

(Macmillan new electronics series, . Introudction to advanced topics)

Macmillan Press, 1993

  • : pbk

Available at  / 8 libraries

Search this Book/Journal

Note

Includes bibliographical references and index

Description and Table of Contents

Volume

ISBN 9780333519219

Description

Recent years have seen the transfer of speech technology from the research laboratory to the market place. A variety of commercial speech products is now available for speech synthesis, automatic speech recognition and speech coding (bandwidth compression). Providing an introduction to the rapidly developing area of speech technology, this book covers the entire spectrum of speech technology, ranging from speech analysis and speech synthesis through to automatic speech recognition and speech coding. The emphasis is mainly on the signal processing aspects of speech and the treatment is primarily descriptive and illustrative, with the mathematical content being kept to a minimum.

Table of Contents

  • The nature of speech - speech production, source-filter model, speech sounds, co-articulation and prosody, waveforms and spectrograms, human auditory system
  • digital speech - sampling, pre-sampling filter, quantization, adaptive differential pulse code modulation (ADPCM), delta modulation
  • parametric speech analysis - pre-emphasis, filterbanks, discrete/fast Fourier transform, cepstral analysis, auto-correlation function, linear predictive analysis, pitch-synchronous analysis
  • feature extraction - short-time energy function, zero-crossing rate, endpoint detection, vector quantization, formant tracking, pitch extraction, phonetic analysis
  • speech synthesis - history, formant sythesizers, linear predictive synthesizers, copy synthesis, phoneme synthesis, concatenation, text-to-speech, articulatory speech synthesis
  • speech coding - subband, transform coding, channel, formant, homomorphic, linear predictive vocoders, vector quantizer coders
  • automatic speech recognition - problems, dynamic time-warping (DTW), hidden Markov models, speaker identification/verification, future trends.
Volume

: pbk ISBN 9780333519226

Description

Providing an introduction to the rapidly developing area of speech technology, this book covers the entire spectrum of speech technology, ranging from speech analysis and speech synthesis through to automatic speech recognition and speech coding. The emphasis is mainly on the signal processing aspects of speech and the treatment is primarily descriptive and illustrative, with the mathematical content being kept to a minimum.

Table of Contents

Series Editor's Foreword - Preface - THE NATURE OF SPEECH - Speech Production - Source-Filter Model - Speech Sounds - Co-articulation and Prosody - Waveforms and Spectrograms - Human Auditory System - DIGITAL SPEECH - Sampling - Pre-sampling Filter - Quantisation - Adaptive Differential Pulse Code Modulation (ADPCM) - Delta Modulation - PARAMETRIC SPEECH ANALYSIS - Pre-Emphasis - Filterbanks - Discrete/Fast Fourier Transform - Cepstral Analysis - Autocorrelation Function - Linear Predictive Analysis - Pitch-Synchronous Analysis - FEATURE EXTRACTION - Short-Time Energy Function - Zero-Crossing Rate - Endpoint Detection - Vector Quantisation - Formant Tracking - Pitch Extration - Phonetic Analysis - SPEECH SYNTHESIS - History - Formant Synthesisers - Linear Predictive Synthesisers - Copy Synthesis - Phoneme Synthesis - Concatenation - Text-to-Speech - Articulatory Speech Synthesis - SPEECH CODING - Subband - Transform Coding - Channel - Formant - Homomorphic - Linear Predicitve Vocoders - Vector Quantiser Coders - AUTOMATIC SPEECH RECOGNITION - Problems - Dynamic Time-Warping (DTW) - Hidden Markov Models - Speaker Identification/Verification - Future Trends - References - Index

by "Nielsen BookData"

Related Books: 1-1 of 1

Details

  • NCID
    BA20763803
  • ISBN
    • 0333519213
    • 0333519221
  • Country Code
    uk
  • Title Language Code
    eng
  • Text Language Code
    eng
  • Place of Publication
    Houndmills
  • Pages/Volumes
    xii, 179 p.
  • Size
    25 cm
  • Classification
  • Parent Bibliography ID
Page Top