Bayesian learning of the SCHMM parameters for speech recognition

Author

Huo, Qiany ; Chan, Chorkin ; Lee, Chin-Hui

Author_Institution

Dept. of Comput. Sci., Hong Kong Univ., Hong Kong

Volume

i

fYear

1994

fDate

19-22 Apr 1994

Abstract

A theoretical framework for Bayesian adaptive learning of semi-continuous HMM parameters is presented. Formulations of MAP estimation of SCHMM parameters are developed. An empirical Bayes method to estimate the hyperparameters of prior densities based on the moment estimate is proposed. Practical issues related to the use of the proposed technique for speaker adaptation application are studied. Effects of various adaptation schemes are examined and their viability is confirmed in a series of comparative experiments using a 26-word English alphabet vocabulary. The proposed method is applicable to other problems in HMM training for speech recognition such as sequential training, context adaptation and parameter smoothing

Keywords

Bayes methods; adaptive systems; estimation theory; hidden Markov models; parameter estimation; speech recognition; Bayesian adaptive learning; English alphabet vocabulary; HMM training; MAP estimation; SCHMM parameters; context adaptation; experiments; hyperparameters estimation; moment estimate; parameter smoothing; semi-continuous HMM parameters; sequential training; speaker adaptation; speech recognition; Bayesian methods; Covariance matrix; Hidden Markov models; Laboratories; Parameter estimation; Smoothing methods; Speech recognition; State estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on

Conference_Location

Adelaide, SA

ISSN

1520-6149

Print_ISBN

0-7803-1775-0

Type

conf

DOI

10.1109/ICASSP.1994.389315

Filename

389315