连续语音端点检测，短时能量、过零率以及谱熵相结合，健壮性好，识别率高。阈值为自适应，可适合情况下语音端点检测。-Continuous speech endpoint detection, short-term energy, zero-crossing rate and spectral entropy, robustness, high recognition rate. The threshold adaptive, may be suitable for the case of voice

熵值越大则每个符号包含的平均信息量越大。有研究发现，在有噪声的语音信号中，语音信号的熵和噪声信号的熵存在着较大的差异，对噪声信号来说在整个频带内分布相对平坦，熵值小，语音信号集中在某些特定频段内，熵值大。因此利用这个差异可以区分噪音段和语音段。(The greater the entropy is, the greater the average information of each symbol is. It is found that, in noisy speech signals, t