A baseline for the transcription of Italian broadcast news

Author

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

Author_Institution

Centro per la Ricerca Sci. e Tecnol., ITC-irst, Trento, Italy

Volume

3

fYear

2000

fDate

2000

Firstpage

1667

Abstract

The paper presents the first achievements in the development of a broadcast news transcription system to be applied for the processing of huge audio archives. In particular, the Italian broadcast news corpus under collection is introduced, and the first implemented baseline system is outlined. The baseline system consists of an audio segmentation module and a speech recognizer featuring a recursive Viterbi beam search, a 64k word lexicon, a tree-based trigram LM representation, and MLLR adaptation. The word error rate of the baseline was 20.9% on planned studio speech and 28.8% on the whole test set

Keywords

audio signal processing; speech recognition; Italian broadcast news transcription; MLLR adaptation; audio archives; audio segmentation module; baseline system; planned studio speech; recursive Viterbi beam search; speech recognizer; tree-based trigram LM representation; word error rate; word lexicon; Acoustic beams; Acoustic signal detection; Acoustic testing; Adaptation model; Decoding; Loudspeakers; Maximum likelihood linear regression; Radio broadcasting; Speech recognition; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on

Conference_Location

Istanbul

ISSN

1520-6149

Print_ISBN

0-7803-6293-4

Type

conf

DOI

10.1109/ICASSP.2000.862070

Filename

862070