Language model adaptation for highly-inflected Slovenian language in comparison to English language /

The language model in this article is meant as an information source of a speech recogniser. The environment to which the recogniser will be put is topic-specific. The idea is to try to adapt the model of general language to the target domain of discourse. We concentrate on feature extraction proces...

Popoln opis

Shranjeno v:
Bibliografske podrobnosti
Main Authors: Sepesy Maučec, Mirjam. (Author), Kačič, Zdravko. (Author)
Format: Book Chapter
Jezik:English
Teme:
Sorodne knjige/članki:Vsebovano v: Proceedings of the Corpus linguistics 2001 conference, Lancaster University (UK), 29 March - 2 April 2001
Oznake: Označite
Brez oznak, prvi označite!
Opis
Izvleček:The language model in this article is meant as an information source of a speech recogniser. The environment to which the recogniser will be put is topic-specific. The idea is to try to adapt the model of general language to the target domain of discourse. We concentrate on feature extraction process devoted to highly-inflected languages. Results of experiments on English and Slovenian corpora are reported.
Fizični opis:Str. 523-530.
Bibliografija:Bibliografija: str. 530.
Abstract.
ISBN:1862201072