A Bayesian prediction approach to robust speech recognition and online environmental learning

被引:5
作者
Chien, JT [1 ]
机构
[1] Natl Cheng Kung Univ, Dept Comp Sci & Informat Engn, Tainan 70101, Taiwan
关键词
Bayesian predictive classification (BPC); online unsupervised learning; speaker adaptation; speech recognition; hidden Markov model;
D O I
10.1016/S0167-6393(01)00032-2
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
A robust speech recognizer is developed to tackle the inevitable mismatch between training and testing environments. Because the realistic environments are uncertain and nonstationary, it is necessary to characterize the uncertainty of speech hidden Markov models (HMMs) for recognition and trace the uncertainty incrementally to catch the newest environmental statistics. In this paper, we develop a new Bayesian predictive classification (BPC) for robust decision and online environmental learning. The BPC decision is adequately established by modeling the uncertainties of both the HMM mean rector and precision matrix using a conjugate prior density. The frame-based predictive distributions using multivariate t distributions and approximate Gaussian distributions are herein exploited. After the recognition, the prior density is pooled with the likelihood of the Current test sentence to generate the reproducible prior density. The hyperparameters of the prior density are accordingly adjusted to meet the newest environments and apply for the recognition of upcoming data. As a result, an efficient online unsupervised learning strategy is developed for HMM-based speech recognition without needing adaptation data. In the experiments, the proposed approach is significantly better than conventional plug-in maximum a posteriori (MAP) decision on the recognition of connected Chinese digits in hands-free car environments. This approach is economical in computation. (C) 2002 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:321 / 334
页数:14
相关论文
共 30 条
[11]  
Huo Q, 1997, IEEE T SPEECH AUDI P, V5, P161, DOI 10.1109/89.554778
[12]   A Bayesian predictive classification approach to robust speech recognition [J].
Huo, Q ;
Lee, CH .
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 2000, 8 (02) :200-204
[13]  
HUO Q, 1998, P INT S CHIN SPOK LA, P31
[14]  
HUO Q, 1997, P EUR C SPEECH COMM, V4, P1847
[15]  
HUO Q, 1997, IEEE P INT C AC SPEE, V2, P1547
[16]   Improving Viterbi Bayesian predictive classification via sequential Bayesian learning in robust speech recognition [J].
Jiang, H ;
Hirose, K ;
Huo, Q .
SPEECH COMMUNICATION, 1999, 28 (04) :313-326
[17]  
JIANG H, 1997, IEEE P INT C AC SPEE, V2, P1551
[18]   ON INFORMATION AND SUFFICIENCY [J].
KULLBACK, S ;
LEIBLER, RA .
ANNALS OF MATHEMATICAL STATISTICS, 1951, 22 (01) :79-86
[19]  
LEE CH, 1999, P WORKSH ROB METH SP, P45
[20]  
Lee JJ, 1998, SYMBIOSIS, V25, P1