发明名称 |
Bubble splitting for compact acoustic modeling |
摘要 |
<p>An improved method is provided for constructing compact acoustic models for use in a speech recognizer. The method includes: partitioning speech data from a plurality of training speakers according to at least one speech related criteria (i.e., vocal tract length); grouping together the partitioned speech data from training speakers having a similar speech characteristic; and training an acoustic bubble model for each group using the speech data within the group. <IMAGE></p> |
申请公布号 |
EP1507255(A2) |
申请公布日期 |
2005.02.16 |
申请号 |
EP20040018341 |
申请日期 |
2004.08.03 |
申请人 |
MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. |
发明人 |
MUTEL, AMBROISE;NGUYEN, PATRICK;RIGAZIO, LUCA |
分类号 |
G10L11/00;G10L15/02;G10L15/06;G10L15/14;(IPC1-7):G10L15/14 |
主分类号 |
G10L11/00 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|