摘 要：本论文主要研究了语音识别的基本原理,对语音识别系统的构成进行分析处理,其中包括预处理、特征参数提取、建立模块库、识别匹配几大部分。预处理又包括语音采样、预加重、加窗（汉明窗）、端点检测；特征提取的参数是梅尔频率倒谱系数MFCC。

该语音系统采用的是动态时间伸缩算法(DTW)，研究对象是特定人的语音识别，并在MATLAB平台上实现。为了进行后续研究，首先使用电脑中的录音系统录制了阿拉伯数字0—9的语音文件，并转化成 “.wav”格式的文件。-Abstract: This thesis mainly studied the basic principle of speech recognition, to analyze the composition of the speech recognition system, including the preprocessing, feature extraction, to set up the module library, identify several most matches. Pretreatment, including speech sampling, pre-emphasis, add window (hamming window), endpoint detection　Feature extraction of MFCC MEL frequency cepstrum coefficient.

The voice system USES a dynamic time scale (DTW) algorithm, the research object is the speaker-dependent speech recognition, and realized in MATLAB platform.To carry out the follow-up study, the first to use the recording in a computer system to record the audio files of Arabic Numbers 0-9, and translated into　. Wav format file.