Invention Grant
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
- Patent Title: Speech models generated using competitive training, asymmetric training, and data boosting
- Patent Title (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
-
Application No.: US11156106Application Date: 2005-06-17
-
Publication No.: US07693713B2Publication Date: 2010-04-06
- Inventor: Xiaodong He , Jian Wu
- Applicant: Xiaodong He , Jian Wu
- Applicant Address: US WA Redmond
- Assignee: Microsoft Corporation
- Current Assignee: Microsoft Corporation
- Current Assignee Address: US WA Redmond
- Agency: Westman, Champlin & Kelly, P.A.
- Agent Joseph R. Kelly
- Main IPC: G10L15/06
- IPC: G10L15/06
Abstract:
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.
Public/Granted literature
- US20060287856A1 Speech models generated using competitive training, asymmetric training, and data boosting Public/Granted day:2006-12-21
Information query
