MMI training for continuous phoneme recognition on the TIMIT database

1 January 1993

conference paper
Published by Institute of Electrical and Electronics Engineers (IEEE)

Vol. 2, 491-494 vol.2
https://doi.org/10.1109/icassp.1993.319349

Abstract

Experiences with a phoneme recognition system for the TIMIT database which uses multiple mixture continuous-density monophone HMMs (hidden Markov models) trained using MMI (maximum mutual information) is reported. A comprehensive set of results are presented comparing the ML (maximum likelihood) and MMI training criteria for both diagonal and full covariance models. These results using simple monophone HMMs show that clear performance gains are achieved by MMI training. These results are comparable with the best reported by others, including those which use context-dependent models. In addition, a number of performance and implementation issues which are crucial to successful MMI training are discussed.

Keywords

This publication has 7 references indexed in Scilit:

Phonetic recognition using hidden Markov models and maximum mutual information training
Published by Institute of Electrical and Electronics Engineers (IEEE) ,2003
An inequality for rational functions with applications to some statistical estimation problems
IEEE Transactions on Information Theory, 1991
Speaker-independent phone recognition using hidden Markov models
IEEE Transactions on Acoustics, Speech, and Signal Processing, 1989
A tutorial on hidden Markov models and selected applications in speech recognition
Proceedings of the IEEE, 1989
On a model-robust training method for speech recognition
IEEE Transactions on Acoustics, Speech, and Signal Processing, 1988
The Acoustic-Modeling Problem in Automatic Speech Recognition.
Published by Defense Technical Information Center (DTIC) ,1987
Optimal solution of a training problem in speech recognition
IEEE Transactions on Acoustics, Speech, and Signal Processing, 1985