Adaptation of polynomial trajectory segment models for large vocabulary speech recognition

Author

Kannan, Ashvin ; Ostendorf, Mari

Author_Institution

Dept. of Electr. & Comput. Eng., Boston Univ., MA, USA

Volume

2

fYear

1997

fDate

21-24 Apr 1997

Firstpage

1411

Abstract

Segment models are a generalization of HMMs that can represent feature dynamics and/or correlation in time. We develop the theory of Bayesian and maximum-likelihood adaptation for a segment model characterized by a polynomial mean trajectory. We show how adaptation parameters can be shared and adaptation detail can be controlled at run-time based on the amount of adaptation data available. Results on the Switchboard corpus show error reductions for unsupervised transcription mode adaptation and supervised batch mode adaptation

Keywords

Bayes methods; correlation methods; hidden Markov models; maximum likelihood estimation; polynomials; speech processing; speech recognition; Bayesian theory; HMM; Switchboard corpus; adaptation data; adaptation parameters; error reductions; feature dynamics; large vocabulary speech recognition; maximum likelihood adaptation; polynomial mean trajectory; polynomial trajectory segment models; supervised batch mode adaptation; time correlation; unsupervised transcription mode adaptation; Bayesian methods; Clustering algorithms; Context modeling; Gaussian processes; Hidden Markov models; Maximum likelihood estimation; Polynomials; Robustness; Speech recognition; Vocabulary;

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.1997.596212

Filename

596212