Language model adaptation using auto-induced semantic structures in a voice search system

Author

Yali Li ; Li, Yali ; Yan, Yonghong

Author_Institution

ThinkIT Lab., Chinese Acad. of Sci., Beijing, China

Volume

3

fYear

2009

fDate

20-22 Nov. 2009

Firstpage

350

Lastpage

353

Abstract

In this paper, we study how to generate in-domain data for statistical language model adaptation in a Chinese voice search dialogue system. Given limited amount of in-domain data, we use unsupervised clustering to induce semantic classes and structures from the first part of test data. These structures are further augmented with domain information to generate large amount of in-domain data. Lastly we test on the second part of test data and get a improvement of speech recognition for 6.2%.

Keywords

natural language processing; pattern clustering; speech recognition; Chinese voice search dialogue system; auto induced semantic structure; in-domain data generation; speech recognition; statistical language model adaptation; unsupervised clustering; voice search system; Acoustics; Adaptation model; Automatic speech recognition; Databases; Hidden Markov models; Laboratories; Natural languages; Speech recognition; Speech synthesis; Testing; semantic class induction; statistic language model adaptation;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Computing and Intelligent Systems, 2009. ICIS 2009. IEEE International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-4754-1

Electronic_ISBN

978-1-4244-4738-1

Type

conf

DOI

10.1109/ICICISYS.2009.5358161

Filename

5358161