Maximum mutual information estimation of hidden Markov model parameters for speech recognition

Bahl, L.R.; Brown, Peter; Souza, P. de; Mercer, R. L.

doi:10.1109/icassp.1986.1169179

articleMar 24, 2005Closed access

Maximum mutual information estimation of hidden Markov model parameters for speech recognition

LBL.R. Bahl PBPeter Brown PDP. de Souza RLR. L. Mercer

IBM Research - Thomas J. Watson Research Center · IBM (United States)

Indexed incrossref

Abstract

A method for estimating the parameters of hidden Markov models of speech is described. Parameter values are chosen to maximize the mutual information between an acoustic observation sequence and the corresponding word sequence. Recognition results are presented comparing this method with maximum likelihood estimation.

Citation impact

828

total citations

FWCI: 55.59
Percentile: 100%
References: 3

Citations per year

Authors

4

Topics & keywords

Topics

Speech Recognition and Synthesis94%

Keywords

Hidden Markov model
Mutual information
Speech recognition
Computer science
Maximum-entropy Markov model
Hidden semi-Markov model
Maximum likelihood
Markov model

UN Sustainable Development Goals

Reduced inequalities

No related works found for this paper.