基于听觉感知的LSA-MMSE改进型语音增强方法  

An Improved LSA-MMSE Enhancement Approach Based on Auditory Perception

作　　者：陈琪[1,2] 郭英[1] 张群[1] 王布宏[1] 

机构地区：[1]空军工程大学电讯工程学院,西安710077 [2]国防科学技术大学电子科学与工程学院,长沙410073

出　　处：《信号处理》2008年第6期1037-1040,共4页Journal of Signal Processing

基　　金：国家自然科学基金(60601016，60672032)资助项目;陕西省自然科学基金(2006F40)资助项目

摘　　要：传统增强方法的增益函数对每个频点都进行估计,必然会引进相对较多的语音失真。为了提高低信噪比下的语音增强效果,提出了一种计算掩蔽概率的方法,得到优化的语音增强方法。基于听觉感知特性,对噪声被掩蔽部分的带噪语音谱和未掩蔽部分采用不同处理方法。增强后的语音可以表示为这两个状态下单独估计的加权和,其中权重与噪声被掩蔽概率有关。通过与Virag的方法、LSA-MMSE估计等方法进行比较,实验结果表明所提的增强方法能在低信噪比下有效地抑制残留噪声的同时保持更小的语音失真。Gain function of traditional enhancement algorithms is to estimate every signal spectral component, therefore, this introduces relatively more speech distortion. To improve the effect of speech enhancement at low signal-to-noise ratio （SNR） ,this paper pro- posed a method of calculating masking threshold and optimal speech enhancement scheme. Based on auditory perception properties ,we took different method for processing noise masked spectrum and noise unmasked spectrum. Then a speech signal estimator is proposed as a weighted sum of the individual estimator in each state,where the weight is related with noise masked probability. Compared with Virag＇ s method and LSA-MMSE estimator, the proposed estimator can suppress the residual noise effectively while keeping smaller speech distortion especially at low SNR.

关 键 词：语音增强 增益函数 掩蔽阈值 噪声被掩蔽概率 对数谱最小均方误差 

分 类 号：TN912.35[电子电信—通信与信息系统] TS802.3[电子电信—信息与通信工程]