发明授权
- 专利标题： Speaker verification system using decision fusion logic
- 专利标题（中）： 扬声器验证系统采用决策融合逻辑
-
申请号： US479012申请日： 1995-06-07
-
公开(公告)号： US5839103A公开(公告)日： 1998-11-17
- 发明人: Richard J. Mammone , Kevin Farrell , Manish Sharma , Devang Naik , Xiaoyu Zhang , Khaled Assaleh , Han-Seng Liou
- 申请人： Richard J. Mammone , Kevin Farrell , Manish Sharma , Devang Naik , Xiaoyu Zhang , Khaled Assaleh , Han-Seng Liou
- 申请人地址： NJ Piscataway
- 专利权人： Rutgers, The State University of New Jersey
- 当前专利权人： Rutgers, The State University of New Jersey
- 当前专利权人地址： NJ Piscataway
- 主分类号： G10L15/10
- IPC分类号： G10L15/10 ; G10L15/02 ; G10L15/08 ; G10L15/20 ; G10L17/00 ; G10L21/02 ; G10L5/00
摘要：
The present invention relates to a pattern recognition system which uses data fusion to combine data from a plurality of extracted features and a plurality of classifiers. Speaker patterns can be accurately verified with the combination of discriminant based and distortion based classifiers. A novel approach using a training set of a "leave one out" data can be used for training the system with a reduced data set. Extracted features can be improved with a pole filtered method for reducing channel effects and an affine transformation for improving the correlation between training and testing data.
公开/授权文献
- US4551980A Hybrid system for generating power 公开/授权日：1985-11-12
信息查询
