发明名称 |
SPEAKER VERIFICATION SYSTEM |
摘要 |
The present invention relates to a pattern recognition system (Fig. 1) which uses data fusion to combine data from a plurality of extracted features (60, 61, 62) and a plurality of classifiers (70, 71, 72). 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 (Figs. 7A, 7B, 7C). Extracted features can be improved with a pole filtered method for reducing channel effects (Fig. 11B) and an affine transformation for improving the correlation between training and testing data (Fig. 14). |
申请公布号 |
WO9641334(A1) |
申请公布日期 |
1996.12.19 |
申请号 |
WO1996US09260 |
申请日期 |
1996.06.06 |
申请人 |
RUTGERS UNIVERSITY, |
发明人 |
MAMMONE, RICHARD, J.;FARRELL, KEVIN;SHARMA, MANISH;DEVANG, NAIK;ZHANG, XIAOYU;ASSALEH, KHALED;LIOU, HAN-SHENG |
分类号 |
G10L15/10;G10L15/02;G10L15/08;G10L15/20;G10L17/00;G10L21/02 |
主分类号 |
G10L15/10 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|