语音端点检测是语音处理和语音识别系统中极其关键的一步.该文首先分析了基于短时能量和平均过零率的语音端点检测方法,这种方法可以在高信噪比的情况下得到较好的识别效果,而随着信噪比的下降识别率也将下降.针对语音波形和信息熵的特点,该文提出了一种基于信息熵的语音端点检测方法.实验结果表明,该方法在信噪比较低的情况下仍能比较准确地检测出语音信号的端点,并且实现简单.“,”Speech endpoint detection is a very important step for speech signals processing and speech recognition system. Firstly, this paper analysis about the endpoint detection methods with short-time energy and zero-crossing-rate, which can locate the endpoint efficiently in the high SNR, however, each of them cannot locate efficiently endpoint for speech in the low SNR. For the features of the voice waveforms and information entropy, a scheme of speech endpoint detection based on information entropy was proposed. Experiments show that the method is simple and effective.