Transcription of broadcast television and radio news: the 1996 ABBOT system

Author

Cook, G.D. ; Kershaw, D.J. ; Christie, J.D.M. ; Seymour, C.W. ; Waterhouse, S.R.

Author_Institution

Dept. of Eng., Cambridge Univ., UK

Volume

2

fYear

1997

fDate

21-24 Apr 1997

Firstpage

723

Abstract

This paper describes the development of the CU-CON system which participated in the 1996 ARPA Hub 4 Evaluations. The system is based on ABBOT, a hybrid connectionist-HMM large vocabulary continuous speech recognition system developed at the Cambridge University Engineering Department by Hochberg et al. (1995). The Hub 4 Evaluation task involves the transcription of broadcast television and radio news programmes. This is an extremely demanding task for a state-of-the-art speech recognition system. Typical programmes include a wide variety of speaking styles and acoustic conditions. These range from read speech recorded in the studio to extemporaneous speech recorded over telephone channels

Keywords

hidden Markov models; neural nets; radio broadcasting; speech recognition; television broadcasting; 1996 ABBOT system; 1996 ARPA Hub 4 Evaluations; CU-CON system; Cambridge University Engineering Department; acoustic conditions; broadcast radio news programmes; broadcast television news programmes; extemporaneous speech; hybrid connectionist-HMM large vocabulary continuous speech recognition system; read speech; speaking styles; studio speech; telephone speech; transcription; Context modeling; Degradation; Hidden Markov models; Loudspeakers; Music; Radio broadcasting; Speech recognition; TV broadcasting; Telephony; 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.1997.596007

Filename

596007