说明：  每组数据为25维，第1维维类别标识，后24维为语音特征信号。然后把四类语音信号何为一组，抽取1500组数据作为训练数据，500组数据作为测试数据，并对数据进行归一化处理。根据语音类别标识设定每组语音信号的期望输出值，如标识类为1时，期望输出向量为[1 0 0 0]。
(Each group of data has 25 dimensions, the first dimension is identified by category, and the last 24 dimensions are speech feature signals.Then, select one group of four speech signals, extract 1500 groups of data as training data and 500 groups of data as test data, and normalize the data.Set the expected output value of each group of speech signals according to the speech category identification. For example, when the identification class is 1, the expected output vector is [1 0 0 0].)