Techniques in Speech Acoustics

Jonathan Harrington and Steve Cassidy
Speech Hearing and Language Research Centre
Macquarie University, Sydney

Published by Kluwer Academic Publishers.

The book can now be ordered from the publisher (Hardbound, ISBN 0-7923-5731-0).

Some of the software included on the accompanying CDROM is described on the Emu Speech Database pages.

Errata

Two problems have been reported on the CDROM included with the book. They are outlined, along with simple fixes, on the errata page.

Table of Contents

• Preface
• Vowel and Consonant Transcriptions
• Contents of the CD-ROM
• 1 The Scope of Speech Acoustics
  • 1.1 What is speech acoustics?
  • 1.2 The variable nature of speech
  • 1.3 Experimental designs in speech acoustics
  • 1.4 Some key areas of research in speech acoustics
• 2 The Physics of Speech
  • 2.1 Speech waveforms
  • 2.2 Frequency analysis
• 3 The Acoustic Theory of Speech Production
  • 3.1 The source-filter decomposition of speech
  • 3.2 The acoustic source in speech production
  • 3.3 The acoustic filter in speech production
  • 3.4 Vocal tract losses
  • 3.5 Radiated sound pressure
  • 3.6 The composite model
• 4 Segmental and Prosodic Cues
  • 4.1 Vowels
  • 4.2 Oral stops
  • 4.3 Nasal consonants
  • 4.4 Fricatives
  • 4.5 Approximants
  • 4.6 Prosody and juncture
• 5 Time-Domain Analysis
  • 5.1 Sampling and quantisation
  • 5.2 Definition of a digital signal
  • 5.3 Simple operations on signals
  • 5.4 Windowing signals
  • 5.5 Some common time-domain parameters
  • 5.6 Convolution and time-domain filtering
• 6 Frequency-Domain Analysis
  • 6.1 Digital sinusoids
  • 6.2 The discrete Fourier transform
  • 6.3 Spectra derived from the DFT
  • 6.4 Some points of procedure in applying a DFT
  • 6.5 Spectral parameterisations
  • 6.6 Frequency-domain filtering
• 7 Digital Formant Synthesis
  • 7.1 Core structure of a formant synthesiser
  • 7.2 Digital considerations
  • 7.3 Periodic excitation
  • 7.4 Formant filter
  • 7.5 Combining the source with the filter
  • 7.6 Parallel structure
• 8 Linear Prediction of Speech
  • 8.1 LPC and its relationship to digital speech
  • 8.2 Techniques for calculating the LPC coefficients
  • 8.3 Analysis of the error signal
  • 8.4 LPC-smoothed spectra and formants
  • 8.5 Area functions and reflection coefficients
  • 8.6 Speech synthesis from LPC-parameters
• 9 Classification of Speech Data
  • 9.1 Speech spaces and distance measures
  • 9.2 Distributions of speech sounds
  • 9.3 Discriminant functions and classification
  • 9.4 Classification experiments
  • 9.5 Classifying signals in time
  • 9.6 Data reduction
• References

Techniques in Speech Acoustics

Enquiries to Steve Cassidy
Last modified: Tue Aug 10 16:30:06 GMT 1999