基于语音端点检测和子空间方法的语音增强算法

利用子空间方法来实现语音的增强,在语音失真和残留噪声之间进行折中处理:既最小化语音失真,同时又使残留噪声保持在一个预先设定的值.传统的子空间法在平稳噪声环境下是有效的,但在非平稳环境下效果却不是很明显,因此利用语音端点检测(VAD)对噪声的协方差进行及时地更新.实验表明,采用基于VAD的子空间方法实现语音增强可以达到很好的效果.

Speech enhancement algorithm based on VAD and subspace method

This paper proposed the subspace-based method to realize the speech enhancement.The authors made a compromise between the residua noise and the speech distortion to minimize the speech distortion while keeping the residual noise at a preset threshold.However,he traditional subspace method is valid in the stationary noise environment,but is invalid in the non-stationary noise environment,so Voice Activity Detection(VAD) method was used to update the noise covariance matrix.The experimental results prove that...