有效抑制短波通信噪声的语音增强算法设计  

Speech enhancement algorithm for effective suppression of noise in short-wave communication

作　　者：李娟[1] 刘珩[1] 蔡乃小[1] 

机构地区：[1]中国农业大学信息与电气工程学院,北京100083

出　　处：《电子测量技术》2010年第3期50-53,共4页Electronic Measurement Technology

摘　　要：针对短波语音信噪比低、噪声复杂的特点,本文结合最佳调整对数谱幅度(OM-LSA)语音估计、改进型最小控制递归平均(IMCRA)噪声估计和谱熵语音激活检测(VAD)硬判决,提出了一种有效抑制短波通信噪声的语音增强算法。实验结果表明,本算法不仅能有效抑制短波噪声对话音的干扰,而且处理后的语音失真度小,语音自然度较高,避免了残留音乐噪声。Considering short-wave radio speech with the characteristic of Low Signal-to-Noise Ratio and complex noise types,this paper presents a speech enhancement algorithm for effective suppression of noise in short-wave communication,which combines the enhanced algorithm based on an optimally-modified log-spectral amplitude（OM-LSA）speech estimator and a improved minima controlled recursive averaging（IMCRA）noise estimator with voice activity detection（VAD）hard-decision based on spectral entropy.Experiment results show that this algorithm not only effectively suppresses the interference on voice from short-wave noise,but also provides the enhanced speech with low distortion and high speech naturalness,and avoids the musical residual noise.

关 键 词：语音增强 短波噪声 OM-LSA IMCRA 谱熵VAD 

分 类 号：TN912.35[电子电信—通信与信息系统]