Dynamic language model adaptation using latent topical information and automatic transcripts

Berlin Chen*

*此作品的通信作者

研究成果: 書貢獻/報告類型會議論文篇章

3 引文 斯高帕斯(Scopus)

摘要

This paper considers dynamic language model adaptation for Mandarin broadcast news recognition. Both contemporary newswire texts and in-domain automatic transcripts were exploited in language model adaptation. A topical mixture model was presented to dynamically explore the long-span latent topical information for language model adaptation. The underlying characteristics and different kinds of model structures were extensively investigated, while their performance was analyzed and verified by comparison with the conventional MAP-based adaptation approaches, which are devoted to extracting the short-span ra-gram information. The fusion of global topical and local contextual information was investigated as well. The speech recognition experiments were conducted on the broadcast news collected in Taiwan. Very promising results in perplexity as well as character error rate reductions were initially obtained.

原文英語
主出版物標題IEEE International Conference on Multimedia and Expo, ICME 2005
頁面97-100
頁數4
DOIs
出版狀態已發佈 - 2005
事件IEEE International Conference on Multimedia and Expo, ICME 2005 - Amsterdam, 荷兰
持續時間: 2005 七月 62005 七月 8

出版系列

名字IEEE International Conference on Multimedia and Expo, ICME 2005
2005

其他

其他IEEE International Conference on Multimedia and Expo, ICME 2005
國家/地區荷兰
城市Amsterdam
期間2005/07/062005/07/08

ASJC Scopus subject areas

  • 工程 (全部)

指紋

深入研究「Dynamic language model adaptation using latent topical information and automatic transcripts」主題。共同形成了獨特的指紋。

引用此