From broadcast news to spontaneous dialogue transcription: portability issues

Author

Bertoldi, N. ; Brugnara, E. ; Cettolo, M. ; Federico, M. ; Giuliani, D.

Author_Institution

ITC-irst, Centro per la Ricerca Scientifica e Tecnologica, Povo di Trento, Italy

Volume

1

fYear

2001

fDate

2001

Firstpage

37

Abstract

Reports on experiments of porting the ITC-irst Italian broadcast news recognition system to two spontaneous dialogue domains. The trade-off between performance and the required amount of task specific data was investigated. Porting was experimented by applying supervised adaptation methods to acoustic and language models. By using two hours of manually transcribed speech, word error rates of 26.0% and 28.4% were achieved by the adapted systems. Two reference systems, developed on a larger training corpus, achieved word error rates of 22.6% and 21.2%, respectively

Keywords

covariance matrices; feature extraction; hidden Markov models; natural languages; speech recognition; statistical analysis; ITC-irst Italian broadcast news recognition system; acoustic models; language models; manually transcribed speech; portability issues; spontaneous dialogue transcription; supervised adaptation methods; task specific data; word error rates; Acoustic measurements; Acoustic testing; Adaptation model; Broadcasting; Costs; Error analysis; Natural languages; Network topology; Speech recognition; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2001. Proceedings. (ICASSP '01). 2001 IEEE International Conference on

Conference_Location

Salt Lake City, UT

ISSN

1520-6149

Print_ISBN

0-7803-7041-4

Type

conf

DOI

10.1109/ICASSP.2001.940761

Filename

940761