语音识别中的语音信号特征提取和选择是关系到语音识别模型性能的一个重要问题。音信号的特征提取是解决时域信号的数字表问题，特征选择则是在多个特征中选择有的特征为后继的模式划分部分提供数据。-Speech recognition in speech signal feature extraction and selection in relation to the performance of speech recognition model of an important issue. Audio signal feature extraction is the solution to the number of time-domain signal that the problem, feature selection is to select a number of features of some of the characteristics of the model for follow-up data into parts.