基于噪声分类和字典选择的语音活动检测  

Voice activity detection based on noise classification and dictionary selection

作　　者：谢怡宁[1] 黄金杰[2] 赵晶[1] 何勇军[1] Xie Yining Huang Jinjie Zhao Jing He Yongjun(School of Computer Science and Technology School of Automation, Harbin University of Science and Technology, Harbin 150080, China)

机构地区：[1]哈尔滨理工大学计算机科学与技术学院,黑龙江哈尔滨150080 [2]哈尔滨理工大学自动化学院,黑龙江哈尔滨150080

出　　处：《华中科技大学学报（自然科学版）》2016年第12期121-126,共6页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金青年基金资助项目(61305001)

摘　　要：为了解决已有语音活动检测方法在噪声条件下性能急剧下降的问题,提出了一种基于稀疏编码的语音活动检测方法.该方法在训练阶段为语音和每种可能的噪声训练字典;在识别阶段首先识别环境噪声类型,然后将语音字典和对应环境噪声字典拼接形成一个大字典用于稀疏分解,最后用语音字典上的稀疏表示实现语音和非语音的判断.该方法一方面引入了噪声类型识别,可以有针对性地选择噪声字典;另一方面引入噪声开集识别机制,能识别新的噪声类型并且为之训练模型.实验结果表明所提方法比传统方法具有更强的噪声鲁棒性.The performance of current voice activity detection （VAD） methods drops substantially in noise condition .To solve this problem ,a new VAD method based on sparse coding was proposed .In the training ,this method learns a dictionary for speech signals and each possible noise ;in the testing , this method first identifies environmental noise types ,and then concatenates the speech dictionary and corresponding environmental noise dictionary to be a large dictionary for sparse decomposition ,and fi-nally uses the representation over speech dictionary to make speech and non-speech classification . Since making using of noise classification ,this method can select noise dictionaries .In addition ,this method makes use of out-set recognition of noises ,which can recognize new noisy and train models for them .Experiments results show that the proposed method is more robust than traditional methods .

关 键 词：语音活动检测 稀疏编码 形态成分分析 K-奇异值分解 噪声鲁棒性 

分 类 号：TP394[自动化与计算机技术—计算机应用技术]