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.
申请公布号
FI117954(B)
申请公布日期
2007.04.30
申请号
FI19970004339
申请日期
1997.11.26
申请人
RUTGERS, THE STATE UNIVESITY OF NEW JERSEY OFFICEOF CORPORATE LIAISON & TECHNOLOGY TRANSFER,THE STATE UNIVERSITY OF