Self-organized language modeling for speech recognition

Self-organized language modeling for speech recognition (1990)

Cached

Download Links

by F. Jelinek , B. Merialdo , S. Roukos , M. Strauss I

Venue:	Readings in Speech Recognition
Citations:	302 - 3 self

Versions

Version History

Metadata Version 1

Datum	Value	Source
TITLE	Self-organized language modeling for speech recognition	INFERENCE
AUTHOR NAME	F. Jelinek	SVM HeaderParse 0.2
AUTHOR AFFIL	IBM Research Division, Thomas J. Watson Research Center,	SVM HeaderParse 0.2
AUTHOR ADDR	Yorktown Heights, NY 10598	SVM HeaderParse 0.2
AUTHOR NAME	B. Merialdo	SVM HeaderParse 0.2
AUTHOR AFFIL	IBM Research Division, Thomas J. Watson Research Center,	SVM HeaderParse 0.2
AUTHOR ADDR	Yorktown Heights, NY 10598	SVM HeaderParse 0.2
AUTHOR NAME	S. Roukos	SVM HeaderParse 0.2
AUTHOR AFFIL	IBM Research Division, Thomas J. Watson Research Center,	SVM HeaderParse 0.2
AUTHOR ADDR	Yorktown Heights, NY 10598	SVM HeaderParse 0.2
AUTHOR NAME	M. Strauss I	SVM HeaderParse 0.2
AUTHOR AFFIL	IBM Research Division, Thomas J. Watson Research Center,	SVM HeaderParse 0.2
AUTHOR ADDR	Yorktown Heights, NY 10598	SVM HeaderParse 0.2
ABSTRACT	In the case of a trlgr~m language model, the proba-bility of the next word conditioned on the previous two words is estimated from a large corpus of text. The re-sulting static trigram language model (STLM) has fixed probabilities that are independent of the document being dictated. To improve the language mode] (LM), one can adapt the probabilities of the trigram language model to match the current document more closely. The partially dictated document provides significant clues about what words ~re more likely to be used next. Of many meth-ods that can be used to adapt the LM, we describe in this paper a simple model based on the trigram frequencies es-timated from the partially dictated document. We call this model ~ cache trigram language model (CTLM) since we are c~chlng the recent history of words. We have found that the CTLM red,aces the perplexity of a dictated doc-ument by 23%. The error rate of a 20,000-word isolated word recognizer decreases by about 5 % at the beginning of a document and by about 24 % after a few hundred words.	SVM HeaderParse 0.2
YEAR	1990	INFERENCE
VENUE	Readings in Speech Recognition	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	450--506	INFERENCE
CITATIONS	3 found	ParsCit 1.0