Improved Modeling and Efficiency for Automatic Transcription of Broadcast News

Improved Modeling and Efficiency for Automatic Transcription of Broadcast News (2001)

Cached

Download Links

by Ananth Sankar , Venkata Ramana Rao Gadde , Andreas Stolcke , Fuliang Weng

Citations:

5 - 0 self

Versions

Version History

Metadata Version 3

User correction supplied by mph

Datum	Value	Source
TITLE	Improved Modeling and Efficiency for Automatic Transcription of Broadcast News	user correction - Legacy Corrections
AUTHOR NAME	Ananth Sankar	user correction
AUTHOR AFFIL		user correction
AUTHOR NAME	Venkata Ramana Rao Gadde	user correction
AUTHOR AFFIL	Speech Technology and Research Laboratory; SRI International, Menlo Park, CA, U.S.A.	user correction
AUTHOR NAME	Andreas Stolcke	user correction
AUTHOR NAME	Fuliang Weng	user correction
AUTHOR AFFIL		user correction
AUTHOR ADDR		user correction
ABSTRACT	Over the last few years, the DARPA-sponsored Hub-4 continuous speech recognition evaluations have advanced speech recognition technology for automatic transcription of broadcast news. In this paper, we report on our research and progress in this domain, with an emphasis on efficient modeling with significantly fewer parameters for faster and more accurate recognition. In the acoustic modeling area, this was achieved through new parameter tying, Gaussian clustering, and mixture weight thresholding schemes. The effectiveness of acoustic adaptation is greatly increased through unsupervised clustering of test data. In language modeling, we explored the use of non-broadcast-news training data, as well as adaptation to topic and speaking styles. We developed an effective and efficient parameter pruning technique for backoff language models that allowed us to cope with ever increasing amounts of training data and expanded N-gram scopes. Finally, we improved our progressive search architecture with more efficient algorithms for lattice generation, compaction, and incorporation of higher-order language models.	user correction
YEAR	2001	user correction - Legacy Corrections
CITATIONS	50 found	ParsCit 1.0