仿听觉掩蔽效应的助听器语音增强算法  

Speech Enhancement Algorithm for Hearing Aids Imitating the Hearing Masking Effect

作　　者：张素贞 唐加能[2] 梁瑞宇[3] 

机构地区：[1]常州轻工技术学院,江苏常州213164 [2]华侨大学工学院,福建泉州362021 [3]南京工程学院通信工程学院,南京211167

出　　处：《电子器件》2016年第3期698-702,共5页Chinese Journal of Electron Devices

基　　金：国家自然基金项目(61375028);江苏省自然基金项目(14KJB510034);泉州市科技计划项目(2014Z103)

摘　　要：为增强听障患者在噪声环境下的语言理解度,提出了一种改进的感知滤波器算法。针对一般感知滤波器会出现被掩蔽音在感知滤波后变为可闻噪声的问题,仿听觉掩蔽效应设计了加权感知滤波器算法。算法在基本感知滤波器设计的基础上,首先采用直接判决法来估计先验信噪比,并基于最佳平滑因子法进行修正。然后基于最小均方差准则获得最优估计值,从而得到感知权重滤波器的增益函数。仿真实验结果表明,该方法能够有效地消除残留噪声,降低噪声对语音的干扰,使得处理后的语音质量有所提高。In order to improve the language comprehension of patients with hearing impairment under the noise envi-ronment,one perception filter algorithm is proposed. For the general perception filter,the masking tone will be changed to audible noise after perception filtering. To solve this problem,one weighted perception filter algorithm for the simulation of the auditory masking effect is designed. Based on the design of basic perception filter ,firstly, the priori SNR is estimated by the direct decision method and revised by the optimal smoothing factor method. The optimal estimate value is obtained with minimum mean square error criterion,and then gain function of perception weighting filter is obtained. The simulation experiment results show that with the proposed method the residual noise can be eliminated effectively and the interference of noise is reduced,so that the quality of the processed speech is improved.

关 键 词：助听器 掩蔽效应 降噪 最小均方差 

分 类 号：TH785.1[机械工程—仪器科学与技术] TB535.1[机械工程—精密仪器及机械]