摘要
提出一种基于样本熵与Mel频率倒谱系数(MFCC)融合的语音情感识别方法。利用支持向量机分别对样本熵统计量与MFCC进行处理,计算其属于高兴、生气、厌烦和恐惧4种情感的概率,采用加法规则和乘法规则对情感概率进行融合,得到识别结果。仿真实验结果表明,该方法的识别率较高。
This paper proposes a method of speech emotion recognition based on fusion of sample entropy and Mel-frequency Cepstral Coefficients(MFCC).Sample entropy statistic and MFCC are modeled with Support Vector Machine(SVM) respectively to obtain the probabilities of happy,angry,bored and afraid.The sum and product rules are used to fuse the probabilities to gain the final decision.Simulation results demonstrate that the recognition rate obtained with the proposed method is high.
出处
《计算机工程》
CAS
CSCD
2012年第7期142-144,共3页
Computer Engineering
基金
国家自然科学基金资助项目(61075008)
关键词
语音情感识别
样本熵
MEL频率倒谱系数
支持向量机
speech emotion recognition
sample entropy
Mel-frequency Cepstral Coefficients(MFCC)
Support Vector Machine(SVM)