在语音增强系统中,为减小语音失真和降低残余噪音,本文推导出一种权系数.这种权系数能保持残余噪音低于噪音掩蔽阈值而语音失真度又较小.Matlab 仿真实验表明,本文提出的语音增强方法,在减少语音失真的同时,能够较好地抑制残余噪音.通过主观听音测试表明语音的自然度也较好. In speech enhancement system, in order to reduce the speech distortion and reduce the residual noise, this paper deduces a weight coefficient, which can keep the residual noise lower than the noise masking threshold and the voice distortion is small.Matlab simulation experiments show that, The speech enhancement method proposed in this paper can restrain the residual noise while reducing the speech distortion, and the subjective listening test shows that the naturalness of the speech is better.