Robust speech recognition based on Viterbi Bayesian predictive classification

Author

Jiang, Hui ; Hirose, Keikichi ; Huo, Qiang

Author_Institution

Dept. of Inf. & Commun. Eng., Tokyo Univ., Japan

Volume

2

fYear

1997

fDate

21-24 Apr 1997

Firstpage

1551

Abstract

In this paper, we investigate a new Bayesian predictive classification (BPC) approach to realize robust speech recognition when there exist mismatches between training and test conditions but no accurate knowledge of the mismatch mechanism is available. A specific approximate BPC algorithm called Viterbi BPC (VBPC) is proposed for both isolated word and continuous speech recognition. The proposed VBPC algorithm is compared with conventional Viterbi decoding algorithm on speaker-independent isolated digit and connected digit string (TIDIGITS) recognition tasks. The experimental results show that VBPC can considerably improve robustness when mismatches exist between training and testing conditions

Keywords

Bayes methods; Viterbi decoding; prediction theory; speech recognition; BPC algorithm; Viterbi Bayesian predictive classification; Viterbi decoding algorithm; continuous speech recognition; isolated word recognition; mismatch mechanism; robust speech recognition; speaker-independent isolated digit string recognition tasks; test conditions; training; Acoustic testing; Automatic speech recognition; Bayesian methods; Decoding; Hidden Markov models; Minimax techniques; Robustness; Signal to noise ratio; Speech recognition; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on

Conference_Location

Munich

ISSN

1520-6149

Print_ISBN

0-8186-7919-0

Type

conf

DOI

10.1109/ICASSP.1997.596247

Filename

596247