发明授权
US07693713B2 Speech models generated using competitive training, asymmetric training, and data boosting
有权
使用竞争性训练，不对称训练和数据提升产生的语音模型
- 专利标题： Speech models generated using competitive training, asymmetric training, and data boosting
- 专利标题（中）： 使用竞争性训练，不对称训练和数据提升产生的语音模型
-
申请号： US11156106申请日： 2005-06-17
-
公开(公告)号： US07693713B2公开(公告)日： 2010-04-06
- 发明人: Xiaodong He , Jian Wu
- 申请人： Xiaodong He , Jian Wu
- 申请人地址： US WA Redmond
- 专利权人： Microsoft Corporation
- 当前专利权人： Microsoft Corporation
- 当前专利权人地址： US WA Redmond
- 代理机构： Westman, Champlin & Kelly, P.A.
- 代理商 Joseph R. Kelly
- 主分类号： G10L15/06
- IPC分类号： G10L15/06
摘要：
Speech models are trained using one or more of three different training systems. They include competitive training which reduces a distance between a recognized result and a true result, data boosting which divides and weights training data, and asymmetric training which trains different model components differently.
公开/授权文献
信息查询
