Statistical Methods for Speech Recognition

Front Cover
MIT Press, 1997 - Computers - 283 pages
3 Reviews

This book reflects decades of important research on the mathematical foundations ofspeech recognition. It focuses on underlying statistical techniques such as hidden Markov models,decision trees, the expectation-maximization algorithm, information theoretic goodness criteria,maximum entropy probability estimation, parameter and data clustering, and smoothing of probabilitydistributions. The author's goal is to present these principles clearly in the simplest setting, toshow the advantages of self-organization from real data, and to enable the reader to apply thetechniques.

  

What people are saying - Write a review

User Review - Flag as inappropriate

This is a fantastic book because it succinctly lays out many of the fundamentals of the field. I find myself repeatedly turning to it because I know it's one place where I'll find a clear and precise explanation of many basic concepts and algorithms that are standard in the field today.
Every time I look at this book it reminds me what a great loss Fred's passing was -- and continues to be -- for all of us in the field today.
Mark Johnson
 

Contents

Chapter
4
Chapter
7
Chapter
12
Chapter
15
Transitions
23
References
37
Basic Language Modeling
45
Chapter 13
50
System
142
Dimensions
158
Language Modeling
166
Chous Method
179
Based on Word Encoding
184
Data
190
Chapter 11
197
Voting
234

History
59
The Viterbi Search
72
State Spaces
86
Recognition
97
Shortcuts
109
Entropy
119
Theorem
126
Adaptation
248
Estimation of Probabilities from Counts
257
Estimate
263
of GoodTuring Estimation
269
Name Index
275
Copyright

Common terms and phrases

Bibliographic information