实时语音端点鲁棒检测

提出了一种适应复杂环境下的高效的实时语音端点检测算法，给出了每帧声信号在滤波中的噪声功率谱的推算方法。先将每帧语音的频谱进行迭代维纳滤波，再将它划分成若干个子带并计算出每个子带的频谱熵，然后把相继若干帧的子带频谱熵经过一组中值滤波器获得每帧的频谱熵，根据频谱熵的值对输入的语音进行分类。实验结果表明，该算法能够有效地区分语音和噪声，可以显著地提高语音识别系统的性能，在不同的噪声环境条件下具有鲁棒性。该算法计算代价小，简单易实现，适合实时语音识别系统的应用。

Real-time voice activity robust detection

This paper proposes an effective real-time voice activity detection algorithm in various noisy environments in the acoustical signal tilting.It makes the iterative methods of estimating the noise power spectrum.The speech spectrum is tilted with iterative Wiener method.The tilting spectrum is divided into several subbands and the spectral entropy of each subband is estimated.Median tilters are applied to a sequence of the subband entropies to obtain the spectral entropy of each frame. The speech/noise classification is based on the spectral entropy.The experimental results show that the proposed algorithm can distinguish speech from noise effectively and improve the performance of automatic speech recognition system significantly. It is proved to be robust under various noisy environments.The algorithm is of low computational complexity which is suitable for real-time speech recognition system application.