Presentation	1998/12/11 Evaluation of a stack decoder on a Japanese Newspaper Dictation Task Mike Schuster,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	This paper describes some of the implementation details of the"Nozomi"stack decoder for LVCSR. The decoder was tasted on a Japanese Newspaper Dictation Task using a 5000 word vocabulary. Using continuous density acoustic models with 2000 and 3000 states trained on the JNAS/ASJ corpora and a 3-gram LM trained on the RWC text corpus, both models peovided by the IPA group[9], it was possible to reach more than 95% word accuracy on the standard test set. With computationally cheap acoustic models we could achieve around 89% accuracy in nearly realtime on a 300 Mhz Pentium II. Using a disk-based LM the memory usage could be optimized to 4 MB in total.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	speech recognition / Japanese newspaper dictation / one-pass stack decoder
Paper #	NLC98-48,SP98-112
Date of Issue

Conference Information
Committee	NLC
Conference Date	1998/12/11(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Natural Language Understanding and Models of Communication (NLC)
Language	ENG
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Evaluation of a stack decoder on a Japanese Newspaper Dictation Task
Sub Title (in English)
Keyword(1)	speech recognition
Keyword(2)	Japanese newspaper dictation
Keyword(3)	one-pass stack decoder
1st Author's Name	Mike Schuster
1st Author's Affiliation	ATR Interpreting Telecommunications Research Laboratories()
Date	1998/12/11
Paper #	NLC98-48,SP98-112
Volume (vol)	vol.98
Number (no)	461
Page	pp.pp.-
#Pages	8
Date of Issue