Improved Discriminative Training Techniques for Large Vocabulary Continuous Speech Recognition (2001)
| Venue: | IEEE ICASSP'01 |
| Citations: | 7 - 0 self |
BibTeX
@INPROCEEDINGS{Povey01improveddiscriminative,
author = {D. Povey and P.C. Woodland},
title = {Improved Discriminative Training Techniques for Large Vocabulary Continuous Speech Recognition},
booktitle = {IEEE ICASSP'01},
year = {2001}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
This paper investigates the use of discriminative training techniques for large vocabulary speech recogntion with training datasets up to 265 hours. Techniques for improving lattice-based Maximum Mutual Information Estimation (MMIE) training are described and compared to Frame Discrimination (FD). An objective function which is an interpolation of MMIE and standard Maximum Likelihood Estimation (MLE) is also discussed. Experimental results on both the Switchboard and North American Business News tasks show that MMIE training can yield significant performance improvements over standard MLE even for the most complex speech recognition problems with very large training sets.
Citations
| 77 | An inequality for rational functions with applications to some statistical estimation problems - Gopalakrishnan, Kanevsky, et al. - 1991 |
| 58 | Large scale discriminative training for speech recognition - Woodland, Povey |
| 33 | Hidden Markov Models, Maximum Mutual Information Estimation and the Speech Recognition Problem - Normandin - 1991 |
| 20 | Frame Discrimination training of HMMs for Large Vocabulary Speech Recognition - Povey, Woodland - 1999 |
| 17 | MMIE training of large vocabulary speech recognition systems - Valtchev, Odell, et al. - 1997 |
| 14 | Discriminative Training of Hidden Markov Models - Kapadia - 1998 |
| 7 | An Investigation of Frame Discrimination for Continuous Speech Recognition - Povey - 1999 |
