这是一本写作较为成功的关于语音语言处理技术的入门级教材，本书可以说打破了两个界限，因而非常值得推荐。首先本书打破了语音和语言处理的界限，在一本教科书中综合讲授了这两个领域的基本概念和方法；二是本书也打破了文理学科的界限，巧妙的选材和灵活的讲授方式，让语言学专业的学生也能免受背景知识的制约，轻易得窥语音语言处理技术的精髓。

导读

Acknowledgements and copyright notices

1 Introduction

 1.1 About this book

 1.2 Purpose of this book

 1.3 Some reasons to use this book

 1.4 What's in the book (and what's not)

 1.5 Computational set-up needed for this book

 1.6 Computational skills that are necessary in order to use the book

 1.7 Free software suggestions

 1.8 Book structure

2 Sounds and numbers

 2.1 Preparatory assignments

 2.2 Solutions

 2.3 Sampling

 2.4 Quantization

 2.5 The sampling theorem

 2.6 Generating a signal

 2.7 Numeric data types

 2.8 The program

 2.9 Structure of a loop

 2.10 Structure of an array

 2.11 Calculating the cosine values

 2.12 Structure of the program

 2.13 Writing the signal to a file

 Chapter summary

 Further Exercises

 Further reading

3 Digital filters and resonators

 3.1 Operations on sequences of numbers

 3.2 A program for calculating RMS amplitude

 3.3 Filtering

 3.4 A program for calculating running means of 4

 3.5 Smoothing over a longer time-window

 3.6 Avoiding the need for long window

 3.7 IIR filters in C

 3.8 Structure of the Klatt formant synthesizer

 Chapter summary

 Exercises

 Further reading

4 Frequency analysis and linear predictive coding

 4.1 Spectral analysis

 4.2 Spectral analysis in C

 4.3 Cepstral analysis

 4.4 Computation of the cepstrum in C

 4.5 Pitch tracking using cepstral analysis

 4.6 Voicing detection

 4.7 f0estimation by the autocorrelation method

 4.8 Linear predictive coding

 4.9 C programs for LPC analysis and resynthesis

 4.10 Trying it out

 4.11 Applications of LPC

 Chapter Summary

 Further exercises

 Further reading

5 Finite-state machines

 5. 1 Some simple examples

 5.2 A more serious example

 5.3 Deterministic and non-deterministic automata

 5.4 Implementation in Prolog

 5.5 Prolog's processing strategy and the treatment of variables

 5.6 Generating strings

 5.7 Three possibly useful applications o{ that idea

 5.8 Another approach to describing finite-state machines

 5.9 Self-loops

 5.10 Finite-state transducers(FSTs)

 5.11 Using finite-state transducers to relate speech to phonemes

 5.12 Finite-state phonology

 5.13 Finite-state syntactic processing

 Chapter summary

 Further exercises

 Further reading

6 Introduction to speech recognition techniques

 6.1 Architectures for speech recognition

 6.2 The pattern-recognition approach

 6.3 Dynamic time warping

 6.4 Applications

 6.5 Sources of variability in speech

 Chapter summary

 Further reading

7 Probabilistic finite-state models

 7.1 Introduction

 7.2 Indeterminacy: n-gram models for part-of-speech tagging

 7.3 Some probability theory for language modelling

 7.4 Markov models

 7.5 Trigram models

 7.6 Incompleteness of the training corpus

 7.7 Part-of-speech model calculations

 7.8 Using HMMs for speech recognition

 7.9 Chomsky's objections to Markov models and some rejoinders

 Chapter summary

 Further reading

8 Parsing

 8.1 Introduction

 8.9 A demo

 8.3 Intuitive parsing

 8.4 Recursive descent parsing

 8.5 The simplest parsing program

 8.6 Difference lists

 8.7 Generating a parse tree

 8.8 Syllabification

 8.9 Other parsing algorithms

 8.10 Chart parsing

 8.11 Depth-first vs. breadth-first search

 8.19 Deterministic parsing, Marcus parsing and minimal commitment parsing

 8.13 Parallel parsing

 Chapter summary

 Further reading

9 Using probabilistie grammars

 9.1 Motivations

 9.2 Probabilistic context-free grammars

 9.3 Estimation of rule probabilities

 9.4 A practical example

 9.5 A limitation of probabilistic context-free grammars

 9.6 Tree adjoining grammars

 9.7 Data-oriented parsing

 Chapter Summary

 Conclusion and suggestions for further reading

Appendix:The American Standard Code for

Information Interchange (ASCII)

Glossary

References

Index