That is, using a soft thresholding to remove noise components from the wavelet coefficients of the voiced and unvoiced speech in noisy speech respectively in a different way, which is not only removing noise but also is preventing the quality degradation
That is, using a soft thresholding to remove noise components from the wavelet coefficients of the voiced and unvoiced speech in noisy speech respectively in a different way, which is not only removing noise but also is preventing the quality degradation
该方法采用软限幅函数对浊音和清音信号的小波变换系数作不同的阈值处理，既抑制了噪声，又减少了语音段信息的损失，提高了信噪比。