基于特征空间分解与融合的语音情感识别  

Speech Emotion Recognition Based on Decomposition of Feature Space and Information Fusion

作　　者：黄程韦[1,2] 金赟[1,2,3] 王青云[1,2] 赵艳[1,2] 赵力[1,2] 

机构地区：[1]东南大学水声信号处理教育部重点实验室,南京210096 [2]东南大学信息科学与工程学院,南京210096 [3]徐州师范大学物理与电子工程学院,徐州221116

出　　处：《信号处理》2010年第6期835-842,共8页Journal of Signal Processing

基　　金：国家自然科学基金项目(60472058;60975017);江苏省自然科学基金项目(BK2008291)

摘　　要：提出了一种语音情感识别中特征空间的优化方法。针对情感类别两两之间的区分度,优化了情感对各自的特征空间,考察了多类分类器分解为两类分类器的方法,采用置信度判决融合的方法进行两类分类器组的重组,实验中比较了单个多类分类器和两类分类器组的识别性能。结果表明,在同等条件下性能提升了8个百分点以上,对多类分类器进行分解,优化每个情感对各自的特征空间,并进行融合的方法适合语音情感识别,对特征空间的优化效果显著。A method of optimizing feature space for speech emotion recognition is proposed.To achieve better classification between each emotion class;feature space of each pair of emotions were optimized respectively;decomposition of multi-class classifier into two-class classifiers was studied;a decision fusion technique was introduced to re-compose the two-class classifier set;recognition results of multi-class classifier and two-class classifier set were compared in a computer experiment.The results show,recognition rates were improved more than 8 percent under identical environments.The method in this paper,decomposition of multi-class classifier,optimizing feature space of each pair of emotions and decomposition using decision fusion algorithm,is suitable for speech emotion recognition and effective in optimization of feature space.

关 键 词：语音情感识别 特征优化 判决融合 

分 类 号：TP391.42[自动化与计算机技术—计算机应用技术]