DocumentCode
353714
Title
Selecting articles from the language model training corpus
Author
Klakow, Dietrich
Author_Institution
Philips GmbH Forschungslab., Aachen, Germany
Volume
3
fYear
2000
fDate
2000
Firstpage
1695
Abstract
The paper suggests the use of a log-likelihood based criterion to select articles from a training corpus that are suitable to reduce perplexity on a specific task defined by a small target corpus. This method is not only efficient as an adaptation technique reducing perplexity by 32% and OOV rate from 4.2% to 2.7% but also as a pruning technique, decreasing the language model size by a factor of 3 at the same time
Keywords
computational linguistics; linguistics; modelling; speech processing; OOV rate; adaptation technique; article selection; language model size; language model training corpus; log-likelihood based criterion; pruning technique; small target corpus; Adaptation model; Educational technology; Entropy; Interpolation; Optimization methods; Testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on
Conference_Location
Istanbul
ISSN
1520-6149
Print_ISBN
0-7803-6293-4
Type
conf
DOI
10.1109/ICASSP.2000.862077
Filename
862077
Link To Document