基于谱减的语音增强算法的改进  

Speech enhancement algorithm based on spectral subtraction

作　　者：李晔[1] 崔慧娟[1] 唐昆[1] 

机构地区：[1]清华大学电子工程系微波与数字通信技术国家重点实验室,北京100084

出　　处：《清华大学学报（自然科学版）》2006年第10期1685-1687,共3页Journal of Tsinghua University(Science and Technology)

基　　金：国家自然科学基金资助项目(60272020)

摘　　要：为降低噪声对语音通信的干扰,提出了一种基于谱减的语音增强算法的改进方法。根据噪声频谱的G auss统计模型修正语音增强过程中噪声频谱的估计方法,利用帧内、帧间约束估计每一个频点的先验信噪比,提出了一种简便的估计语音在每一个频点出现的概率的方法,得出了修正的语音增强算法。在白噪声以及坦克噪声环境中分别对算法的性能进行测试。实验结果表明:在几乎不损伤语音清晰度的前提下,该算法使“音乐噪声”得到了更好的抑制,同时信噪比提高了约8 dB以上。A speech enhancement algorithm was developed based on spectral subtraction to reduce the disturbances of noise on speech communications. The algorithm uses a Gaussian statistical model to revise the noise spectrum estimate for the speech enhancement. Then the algorithm estimates the prior signal to noise ratio of each frequency bin by utilizing intra-frame as well as inter-frame restriction. The algorithm then uses a simple method to compute the presence probability of speech in each frequency bin to enhance the speech signal. The algorithm performance was tested in both white and tank noise environments. The results show that the algorithm more effectively eliminates musical noise and improves the signal to noise rates by 8 dB without significantly impairing the speech intelligibility.

关 键 词：语音信号处理 语音增强 端点检测 信噪比 

分 类 号：TN912.3[电子电信—通信与信息系统]