Discriminative training of Acoustic Models in a Segment-Based Speech Recognizer (2000)
by
I. Lee Hetherington
,
Eric D. Sandness
,
Eric D. Sandness
| Citations: | 1 - 0 self |
BibTeX
@MISC{Hetherington00discriminativetraining,
author = {I. Lee Hetherington and Eric D. Sandness and Eric D. Sandness},
title = {Discriminative training of Acoustic Models in a Segment-Based Speech Recognizer},
year = {2000}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
This thesis explores the use of discriminative training to improve acoustic modeling in a segment-based speech recognizer. In contrast with the more commonly used Maximum Likelihood training, discriminative training considers the likelihoods of competing classes when determining the parameters for a given class's model. Thus, discriminative training works directly to minimize the number of errors made in the recognition of the training data.
