Getting More Mileage from Web Text Sources for Conversational Speech Language Modeling using Class-Dependent Mixtures (2003)
Cached
Download Links
- [crow.ee.washington.edu]
- [ssli.ee.washington.edu]
- [acl.ldc.upenn.edu]
- CiteULike
- DBLP
Other Repositories/Bibliography
by
Ivan Bulyko
,
Mari Ostendorf
| Venue: | Proc. HLT-NAACL 2003 |
| Citations: | 36 - 8 self |
BibTeX
@INPROCEEDINGS{Bulyko03gettingmore,
author = {Ivan Bulyko and Mari Ostendorf},
title = {Getting More Mileage from Web Text Sources for Conversational Speech Language Modeling using Class-Dependent Mixtures},
booktitle = {Proc. HLT-NAACL 2003},
year = {2003},
pages = {7--9}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Sources of training data suitable for language modeling of conversational speech are limited. In this paper, we show how training data can be supplemented with text from the web filtered to match the style and/or topic of the target recognition task, but also that it is possible to get bigger performance gains from the data by using class-dependent interpolation of N-grams.
