Connectionist Speech Recognition: A Hybrid Approach

✍ Scribed by Hervé A. Bourlard, Nelson Morgan (auth.)

• Publisher: Springer US
• Year: 1994
• Tongue: English
• Leaves: 328
• Series: The Springer International Series in Engineering and Computer Science 247
• Edition: 1
• Category: Library

✦ Synopsis

Connectionist Speech Recognition: A Hybrid Approach describes the theory and implementation of a method to incorporate neural network approaches into state of the art continuous speech recognition systems based on hidden Markov models (HMMs) to improve their performance. In this framework, neural networks (and in particular, multilayer perceptrons or MLPs) have been restricted to well-defined subtasks of the whole system, i.e. HMM emission probability estimation and feature extraction.
The book describes a successful five-year international collaboration between the authors. The lessons learned form a case study that demonstrates how hybrid systems can be developed to combine neural networks with more traditional statistical approaches. The book illustrates both the advantages and limitations of neural networks in the framework of a statistical systems.
Using standard databases and comparison with some conventional approaches, it is shown that MLP probability estimation can improve recognition performance. Other approaches are discussed, though there is no such unequivocal experimental result for these methods.
Connectionist Speech Recognition is of use to anyone intending to use neural networks for speech recognition or within the framework provided by an existing successful statistical approach. This includes research and development groups working in the field of speech recognition, both with standard and neural network approaches, as well as other pattern recognition and/or neural network researchers. The book is also suitable as a text for advanced courses on neural networks or speech processing.

✦ Table of Contents

Front Matter....Pages i-xxviii
Front Matter....Pages 2-2
Introduction....Pages 3-13
Statistical Pattern Classification....Pages 15-25
Hidden Markov Models....Pages 27-58
Multilayer Perceptrons....Pages 59-80
Front Matter....Pages 81-81
Speech Recognition Using ANNs....Pages 83-114
Statistical Inference in MLPs....Pages 115-153
The Hybrid HMM/MLP Approach....Pages 155-183
Experimental Systems....Pages 185-200
Context-Dependent MLPs....Pages 201-213
System Tradeoffs....Pages 215-221
Training Hardware and Software....Pages 223-230
Front Matter....Pages 231-231
Cross-validation in MLP Training....Pages 233-241
HMM/MLP and Predictive Models....Pages 243-252
Feature Extraction by MLP....Pages 253-263
Front Matter....Pages 265-265
Final System Overview....Pages 267-274
Conclusions....Pages 275-280
Back Matter....Pages 281-313

✦ Subjects

Circuits and Systems;Statistical Physics, Dynamical Systems and Complexity;Signal, Image and Speech Processing;Electrical Engineering