A Method for speech dereverberation based on an image deblurring algorithm using the prior of speech magnitude gradient distribution in the time-frequency domain

초록

We propose a speech dereverberation method in the timefrequency domain, based on an image deblurring algorithm. A reverberant speech magnitude can be modeled as a convolution of a clean speech with a reverberation filter in time-frequency domain. Then, dereverberation problem can be regarded as that of image deblurring. Therefore, the proposed method estimates the clean speech magnitude in the time-frequency domain by using the fast image deconvolution method with priors on sparsity of the clean speech magnitude gradient and exponentially decaying property of reverberation filters along the time axis. Then, scaling the reverberation speech magnitude by a mask obtained from the estimated clean one performs dereverberation. Experimental results show that the described method can enhance speech. © 2015 ACM.

키워드