On using entropy information to improve posterior probability-based confidence measures

Tzan Hwei Chen*, Berlin Chen, Hsin Min Wang

*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

4 引文 斯高帕斯(Scopus)

摘要

In this paper, we propose a novel approach that reduces the confidence error rate of traditional posterior probability-based confidence measures in large vocabulary continuous speech recognition systems. The method enhances the discriminability of confidence measures by applying entropy information to the posterior probability-based confidence measures of word hypotheses. The experiments conducted on the Chinese Mandarin broadcast news database MATBN show that entropy-based confidence measures outperform traditional posterior probability-based confidence measures. The relative reductions in the confidence error rate are 14.11% and 9.17% for experiments conducted on field reporter speech and interviewee speech, respectively.

原文英語
主出版物標題Chinese Spoken Language Processing - 5th International Symposium, ISCSLP 2006, Proceedings
頁面454-463
頁數10
DOIs
出版狀態已發佈 - 2006 十二月 1
事件5th International Symposium on Chinese Spoken Language Processing, ISCSLP 2006 - Singapore, 新加坡
持續時間: 2006 十二月 132006 十二月 16

出版系列

名字Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
4274 LNAI
ISSN(列印)0302-9743
ISSN(電子)1611-3349

其他

其他5th International Symposium on Chinese Spoken Language Processing, ISCSLP 2006
國家/地區新加坡
城市Singapore
期間2006/12/132006/12/16

ASJC Scopus subject areas

  • 理論電腦科學
  • 電腦科學(全部)

指紋

深入研究「On using entropy information to improve posterior probability-based confidence measures」主題。共同形成了獨特的指紋。

引用此