- ISBN10: 0123865603
Talker Variability in Speech Processing
- List Price: $65.95
- Binding: Hardcover
- Edition: 1
- Publisher: Academic Pr
- Publish date: 01/01/1997
Talker Variability in Speech Processing is a document of current research in speaker variability: the study of the differences in people's voices. It combines new research in speech perception, auditory word recognition, and automatic computer speech recognition systems and demonstrates how linguists, psychologists, and engineers are developing a unified approach to variability issues.
"The editors have assembled an outstanding sample of research on the linguistic, perceptual, technological and clinical impact of differences among talkers. Their timely, superb collection presents a diversity of viewpoints and methods on a topic of increasing importance. This is a very useful book". -- Robert Remez, Barnard College, Columbia University, New York, New York
"This book provides a valuable overview of important problems of speaker variability that affect virtually all aspects of speech research". -- Terrance M. Neary, University of Alberta, Edmonton, Canada
Unique in its approach, Talker Variability in Speech Processing embraces the differences in speech patterns without treating them as unwanted variables. The editors take on the difficult task of converting the mapping of speech patterns into mental representations. They cover theories of perception and cognition, issues in clinical speech pathology, and the practical concerns of speech technology. A radical departure from traditional approaches to speech processing, this text will strike a major chord for those surrounded by the dissonance of speech perception and language processing issues.
-- Contains many superb illustrations
-- Explores talker variability, one of the central problems in automatic computer speechrecognition
-- Provides broad views of the problem and various attempts to solve it
-- Essential for researchers working on speech perception, the higher levels of language processing, and automatic speech recognition