The IBM 2011 GALE Arabic speech transcription system

Author

Mangu, Lidia ; Kuo, Hong-Kwang ; Chu, Stephen ; Kingsbury, Brian ; Saon, George ; Soltau, Hagen ; Biadsy, Fadi

Author_Institution

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

fYear

2011

fDate

11-15 Dec. 2011

Firstpage

272

Lastpage

277

Abstract

We describe the Arabic broadcast transcription system fielded by IBM in the GALE Phase 5 machine translation evaluation. Key advances over our Phase 4 system include a new Bayesian Sensing HMM acoustic model; multistream neural network features; a MADA vowelized acoustic model; and the use of a variety of language model techniques with significant additive gains. These advances were instrumental in achieving a word error rate of 7.4% on the Phase 5 evaluation set, and an absolute improvement of 0.9% word error rate over our 2009 system on the unsequestered Phase 4 evaluation data.

Keywords

Bayes methods; hidden Markov models; neural nets; speech processing; Bayesian sensing HMM acoustic model; GALE Phase 5 machine translation evaluation; IBM 2011 GALE Arabic speech transcription system; MADA vowelized acoustic model; language model techniques; multistream neural network features; phase 4 system; unsequestered phase 4 evaluation data; Acoustics; Computational modeling; Dictionaries; Hidden Markov models; Lattices; Training; Transforms; large vocabulary speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Automatic Speech Recognition and Understanding (ASRU), 2011 IEEE Workshop on

Conference_Location

Waikoloa, HI

Print_ISBN

978-1-4673-0365-1

Electronic_ISBN

978-1-4673-0366-8

Type

conf

DOI

10.1109/ASRU.2011.6163943

Filename

6163943