发明名称 |
Utterance verification using word based minimum verification error training for recognizing a keyword string |
摘要 |
<p>A speech recognition method and apparatus which has a first stage to provide keyword hypotheses and a second stage to provide testing of those hypotheses by utterance verification. The utterance verification used has three separate models for each word: one keyword verification model, one misrecognition verification model, and one non-keyword verification model. Further, all three are developed independently of the recognizer keyword models. Because of this independence, the three verification models can be iteratively trained using existing speech data bases to jointly provide a minimum amount of verification errors. <IMAGE></p> |
申请公布号 |
EP0758781(A2) |
申请公布日期 |
1997.02.19 |
申请号 |
EP19960305618 |
申请日期 |
1996.07.31 |
申请人 |
AT&T IPM CORP. |
发明人 |
SETLUR, ANAND RANGASWAMY;SUKKAR, RAFID ANTOON |
分类号 |
G10L15/00;G10L15/14;(IPC1-7):G10L3/00 |
主分类号 |
G10L15/00 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|