New in this version: Support for multi-class pattern recognition using maxwins, pairwise [4] and DAG-SVM [5] algorithms. A model selection criterion (the xi-alpha bound [6,7] on the leave-one-out cross-validation error). -New in this version

基于单类SVM的说话人语音分割的算法源程序-Based on one-class SVM speaker segmentation source code