基于语音活动性检测的维纳滤波处理

Wiener Filtering Based on Speech Activity Detection

信号处理和信号传输是电子通信技术主要的两个方面。通常纯净的语音经过传输将会被噪声所污染,使得可懂度、识别度降低。而信息处理则是要以最大限度来处理带噪语音,恢复原始语音信号。这个问题也就是滤波问题。针对语音活动性检测可以分离带噪语音中噪声和原始信号的特点,对加噪语音进行处理,以处理前后的信噪比和可懂度作为评判标准。通过实验仿真对实验结果进行评价和分析。结果表明,该方法在一定条件下可以达到预期滤波效果。

Signal processing and signal transmission are two main aspects of electronic communication technology.Usually the pure voice will be polluted by noise after transmission,which will reduce the intelligibility and recognition.Information processing is to process the noisy speech to the maximum extent and restore the original speech signal.This problem is also called filtering prob lem.In view of the fact that speech activity detection can separate noise from original signal in noisy speech.The noise-added speech is processed,and the signal-to-noise ratio and intelligibility before and after processing are taken as the criteria.Evaluation and analysis of experimental results through experimental simulation.The results show that the proposed method can achieve the de sired filtering effect under certain conditions.