一种基于自适应滤波的语音降噪方法研究

分析和研究自适应滤波和小波变换法的原理及方法，提出了一种新的综合使用自适应滤波和小波变换法的语音降噪方法。该方法首先用仿生小波变换法对带噪声的语音信号进行小波分解，将小渡变换法分离出来的噪声信号作为自适应滤波器的输入。最后选择用最小均方误差（LMS）的自适应算法对带噪声语音信号进行降噪处理，实现了信噪分离，去除语音信号中的噪声信号。实验结果表明，该方法对语音信号有较为明显的降噪效果。

Speech de-noising method based on adaptive filtering

The principle and method of the adaptive filtering and the wavelet transform are analyzed and researched. A new method on speech de-nosing with the adaptive filtering and wavelet transform is proposed. The wavelet decomposition of the speech signal with noise is conducted with the method of the bionic wavelet transform. The noise signal separated by the wavelet transform method is taken as the input signal of the adaptive filter. The adaptive filtering algorithm with least-meansquare （LMS） is adopted to implement de-noising processing, realize signal-noise decomposition and eliminate the noise signal in the speech signal. The experimental results show that the method has a prominent effect on the sneech signal de-noising.