一种语音增强算法的研究

The Study of Speech Enhancement Algorithm Based on Subspace

采用基于子空间的方法,对语音信号进行增强处理,该算法提供了在语音信号失真和残留噪声之间进行控制的机制。克服了以往语音增强算法中对语音信号的噪声特点的限制,可以对混有加性白噪声、有色噪声和音乐噪声的语音信号进行增强处理。

It is developed a speech enhancement algorithm based on subspace. The algorithm provides a way to control the speech distortion and residual noise, It can be applied to process the speech with white noise, color noise and music noise.