Recent improvements to IBM´s speech recognition system for automatic transcription of broadcast news

Author

Chen, S.S. ; Eide, E.M. ; Gales, M.J.F. ; Gopinath, R.A. ; Kanevsky, D. ; Olsen, P.

Author_Institution

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

Volume

1

fYear

1999

fDate

15-19 Mar 1999

Firstpage

37

Abstract

We describe extensions and improvements to IBM´s system for automatic transcription of broadcast news. The speech recognizer uses a total of 160 hours of acoustic training data, 80 hours more than for the system described in Chen et al. (1998). In addition to improvements obtained in 1997 we made a number of changes and algorithmic enhancements. Among these were changing the acoustic vocabulary, reducing the number of phonemes, insertion of short pauses, mixture models consisting of non-Gaussian components, pronunciation networks, factor analysis (FACILT) and Bayesian information criteria (BIC) applied to choosing the number of components in a Gaussian mixture model. The models were combined in a single system using NIST´s script voting machine known as rover (Fiscus 1997)

Keywords

Bayes methods; Gaussian processes; speech recognition; BIC; Bayesian information criteria; FACILT; Gaussian mixture model; IBMs speech recognition system; NIST script voting machine; acoustic training; acoustic vocabulary; algorithmic enhancements; automatic transcription; broadcast news; factor analysis; mixture models; nonGaussian components; phonemes; pronunciation networks; rover; short pauses; speech recognizer; Bayesian methods; Broadcasting; Hidden Markov models; Information analysis; Speech analysis; Speech enhancement; Speech recognition; Telephony; Vocabulary; Voting;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on

Conference_Location

Phoenix, AZ

ISSN

1520-6149

Print_ISBN

0-7803-5041-3

Type

conf

DOI

10.1109/ICASSP.1999.758056

Filename

758056