Referenced Items are not available for this document.
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.
Published in:
Automatic Speech Recognition and Understanding (ASRU), 2011 IEEE Workshop on
Date of Conference: 11-15 Dec. 2011
- Page(s):
- 272 - 277
- E-ISBN :
- 978-1-4673-0366-8
- Print ISBN:
- 978-1-4673-0365-1
- INSPEC Accession Number:
- 12577654
- Conference Location :
- Waikoloa, HI
- Digital Object Identifier :
- 10.1109/ASRU.2011.6163943
- Product Type:
- Conference Publications
About IEEE Xplore | Contact | Help | Terms of Use | Nondiscrimination Policy | Site Map | Privacy & Opting Out of Cookies
A not-for-profit organization, IEEE is the world's largest professional association for the advancement of technology.
© Copyright 2013 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.
