A speech understanding framework that uses multiple language models and multiple understanding models

Masaki Katsumaru, Mikio Nakano, Kazunori Komatani, Kotaro Funakoshi, Tetsuya Ogata, Hiroshi G. Okuno

研究成果: Conference contribution

3 被引用数 (Scopus)

抄録

The optimal combination of language model (LM) and language understanding model (LUM) varies depending on available training data and utterances to be handled. Usually, a lot of effort and time are needed to find the optimal combination. Instead, we have designed and developed a new framework that uses multiple LMs and LUMs to improve speech understanding accuracy under various situations. As one implementation of the framework, we have developed a method for selecting the most appropriate speech understanding result from several candidates. We use two LMs and three LUMs, and thus obtain six combinations of them. We empirically show that our method improves speech understanding accuracy. The performance of the oracle selection suggests further potential improvements in our system.

本文言語English
ホスト出版物のタイトルNAACL-HLT 2009 - Human Language Technologies
ホスト出版物のサブタイトル2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Short Papers
編集者Mari Ostendorf, Michael Collins, Shri Narayanan, Douglas W. Oard, Lucy Vanderwende
出版社Association for Computational Linguistics (ACL)
ページ133-136
ページ数4
ISBN(電子版)9781932432428
出版ステータスPublished - 2009
外部発表はい
イベント2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2009 - Boulder, United States
継続期間: 2009 5月 312009 6月 5

出版物シリーズ

名前NAACL-HLT 2009 - Human Language Technologies: 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Short Papers

Conference

Conference2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2009
国/地域United States
CityBoulder
Period09/5/3109/6/5

ASJC Scopus subject areas

  • 言語および言語学
  • 言語学および言語

フィンガープリント

「A speech understanding framework that uses multiple language models and multiple understanding models」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル