一种用于说话人辨认的概率神经网络的MCE训练算法。

两类数据分类器，基于简单高斯模型，并依据MCE最小分类错误率准则进行迭代参数估计后，在对测试数据进行分类-Two types of data classification, based on a simple Gaussian model, and based on the minimum classification error rate criterion MCE iterative parameter estimation after the test data classificatio

信号统计建模，MCE算法,EM算法，模式识别-Signal statistical modeling, MCE algorithm, EM algorithm, pattern recognition

用MCE算法训练分类器，用Train.txt里的数据进行训练，用Test.txt的数据进行性能检测-Use MCE training algorithm of MCE to train the classifier, use the training data in Train.txt to train the classifier and use the data in the Test.txt to test the performance of the classifier