说话人识别代码,端点检测,预加重,MFCC..采用的模型是神经网络~-Speaker identification code, endpoint detection, pre-emphasis, MFCC .. neural network model is adopted ~