基于非因果先验信噪比估计的语音增强改进算法

摘要

为提高MMSE-LSA语音增强算法在低信噪比下的语音增强效果,提出一种改进的MMSE-LSA算法.该算法采用非因果先验信噪比估计法来估计先验信噪比,并引入无语音概率的思想,对增益函数进行改进.实验结果表明,相比传统MMSE-LSA算法,改进算法能更好地抑制残留噪声,提高语音的信噪比,增强效果更好.%To improve the effect of MMSE-LSA speech enhancement algorithm at low signal-to-noise ratio(SNR), thisrnpaper proposes an improved MMSE-LSA algorithm, which uses noncausal estimation for a priori SNR, and introduces arnknowledge of speech absence probability so as to modify gain function from MMSE-LSA. Experimental results showrnthat the proposed algorithm is better than traditional MMSE-LSA algorithm in residual noise suppression and SNRrnimprovement.