Discriminative training of CDHMMs for maximum relative separation margin | IEEE Conference Publication | IEEE Xplore

Discriminative training of CDHMMs for maximum relative separation margin


Abstract:

In this paper, we propose a new discriminative training method for estimating CDHMM (continuous density hidden Markov model) in speech recognition, based on the principle...Show More

First Page of the Article

Abstract:

In this paper, we propose a new discriminative training method for estimating CDHMM (continuous density hidden Markov model) in speech recognition, based on the principle of maximizing the minimum relative multi-class separation margin. We show that the new training criterion can be formulated as a standard constrained minimax optimization problem. Then we show that the optimization problem can be solved by a GPD (generalized probabilistic descent) algorithm. Experimental results on E-set and Alphabet tasks (ISOLET database) showed that the new training criterion can achieve significant (up to 21%) error rate reduction over the popular MCE (minimum classification error) training method.
Date of Conference: 23-23 March 2005
Date Added to IEEE Xplore: 09 May 2005
Print ISBN:0-7803-8874-7

ISSN Information:

Conference Location: Philadelphia, PA, USA

First Page of the Article


Contact IEEE to Subscribe

References

References is not available for this document.