发明名称 |
Speech recognition method using speaker cluster models |
摘要 |
In speaker-independent speech recognition, between-speaker variability is one of the major resources of recognition errors. A speaker cluster model is used to manage recognition problems caused by between-speaker variability. In the training phase, the score function is used as a discriminative function. The parameters of at least two cluster-dependent models are adjusted through a discriminative training method to improve performance of the speech recognition.
|
申请公布号 |
US6567776(B1) |
申请公布日期 |
2003.05.20 |
申请号 |
US20000542844 |
申请日期 |
2000.04.04 |
申请人 |
INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE |
发明人 |
CHANG SEN-CHIA;CHIEN SHIH-CHIEH;PENWU CHUNG-MOU |
分类号 |
G10L15/06;G10L15/14;(IPC1-7):G10L15/06 |
主分类号 |
G10L15/06 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|