A high-performance Cantonese keyword search system

Author

Kingsbury, Brian ; Jia Cui ; Xiaodong Cui ; Gales, Mark J.F. ; Knill, Kate ; Mamou, Jonathan ; Mangu, Lidia ; Nolden, David ; Picheny, Michael ; Ramabhadran, Bhuvana ; Schluter, Ralf ; Sethy, Abhinav ; Woodland, Philip C.

Author_Institution

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

fYear

2013

Firstpage

8277

Lastpage

8281

Abstract

We present a system for keyword search on Cantonese conversational telephony audio, collected for the IARPA Babel program, that achieves good performance by combining postings lists produced by diverse speech recognition systems from three different research groups. We describe the keyword search task, the data on which the work was done, four different speech recognition systems, and our approach to system combination for keyword search. We show that the combination of four systems outperforms the best single system by 7%, achieving an actual term-weighted value of 0.517.

Keywords

natural language processing; speech recognition; telephony; Cantonese conversational telephony audio; Cantonese keyword search system; IARPA Babel program; diverse speech recognition; term-weighted value; Acoustics; Decoding; Indexes; Keyword search; Speech; Speech recognition; Training; deep learning; keyword search; spoken term detection; system combination;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on

Conference_Location

Vancouver, BC

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2013.6639279

Filename

6639279