Retrieval of broadcast news documents with the THISL system

Author

Abberley, Dave ; Renals, Steve ; Cook, Gary

Author_Institution

Dept. of Comput. Sci., Sheffield Univ., UK

Volume

6

fYear

1998

fDate

12-15 May 1998

Firstpage

3781

Abstract

This paper describes a spoken document retrieval system, combining the ABBOT large vocabulary continuous speech recognition (LVCSR) system developed by Cambridge University, Sheffield University and SoftSound, and the PRISE information retrieval engine developed by NIST. The system was constructed to enable us to participate in the TREC 6 Spoken Document Retrieval experimental evaluation. Our key aims in this work were to produce a complete system for the SDR task, to investigate the effect of a word error rate of 30-50% on retrieval performance and to investigate the integration of LVCSR and word spotting in a retrieval task

Keywords

broadcasting; error statistics; information retrieval system evaluation; speech recognition; ABBOT; Cambridge University; LVCSR; NIST; PRISE information retrieval engine; Sheffield University; SoftSound; THISL system; TREC 6 Spoken Document Retrieval; broadcast news documents retrieval; experimental evaluation; large vocabulary continuous speech recognition; retrieval performance; spoken document retrieval system; word error rate; word spotting; Broadcasting; Computer science; Decoding; Error analysis; Hidden Markov models; Indexing; Information retrieval; NIST; Speech recognition; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on

Conference_Location

Seattle, WA

ISSN

1520-6149

Print_ISBN

0-7803-4428-6

Type

conf

DOI

10.1109/ICASSP.1998.679707

Filename

679707