On a model-robust training method for speech recognition

Author

Nádas, Arthur ; Nahamoo, David ; Picheny, Michael A.

Author_Institution

IBM Thomas J. Watson Res. Center, Yorktown Heights, NY, USA

Volume

36

Issue

9

fYear

1988

fDate

9/1/1988 12:00:00 AM

Firstpage

1432

Lastpage

1436

Abstract

Training methods for designing better decoders are compared. The training problem is considered as a statistical parameter estimation problem. In particular, the conditional maximum likelihood estimate (CMLE), which estimates the parameter values that maximize the conditional probability of words given acoustics during training, is compared to the maximum-likelihood estimate, which is obtained by maximizing the joint probability of the words and acoustics. For minimizing the decoding error rate of the (optimal) maximum a posteriori probability (MAP) decoder, it is shown that the CMLE (or maximum mutual information estimate, MMIE) may be preferable when the model is incorrect. In this sense, the CMLE/MMIE appears more robust than the MLE

Keywords

decoding; errors; speech recognition; conditional maximum likelihood estimate; decoders; decoding error rate; model-robust training method; speech recognition; statistical parameter estimation problem; Acoustics; Design methodology; Error analysis; Maximum likelihood decoding; Maximum likelihood estimation; Mutual information; Parameter estimation; Probability; Robustness; Speech recognition;

fLanguage

English

Journal_Title

Acoustics, Speech and Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

0096-3518

Type

jour

DOI

10.1109/29.90371

Filename

90371