首页 | 本学科首页   官方微博 | 高级检索  
     

结合音质特征和韵律特征的语音情感识别
引用本文:张石清,赵知劲,雷必成,杨广映. 结合音质特征和韵律特征的语音情感识别[J]. 电路与系统学报, 2009, 14(4)
作者姓名:张石清  赵知劲  雷必成  杨广映
作者单位:1. 杭州电子科技大学,通信工程学院,浙江,杭州,310018;台州学院,物理与电子工程学院,浙江,临海,317000
2. 杭州电子科技大学,通信工程学院,浙江,杭州,310018
3. 台州学院,物理与电子工程学院,浙江,临海,317000
基金项目:浙江省教育厅高校青年教师资助 
摘    要:为了提高语音情感的正确识别率,在情感语音韵律特征的基础上,提出情感语音音质特征的提取.结合音质特征参数和韵律特征参数,采用支持向量机分类器实现汉语普通话生气、高兴、悲伤和惊奇四种主要情感类型语音的情感识别.实验结果表明,语音音质特征参数和韵律特征参数相结合取得的情感平均正确识别率为88.1%,比单独使用韵律特征参数高出6%.可见,语音音质特征是一种较有效的情感特征参数.

关 键 词:韵律特征  音质特征  支持向量机  语音情感识别

Speech emotion recognition by combining voice quality and prosody features
ZHANG Shi-qing,,ZHAO Zhi-jin,LEI Bi-cheng,YANG Guang-ying. Speech emotion recognition by combining voice quality and prosody features[J]. Journal of Circuits and Systems, 2009, 14(4)
Authors:ZHANG Shi-qing    ZHAO Zhi-jin  LEI Bi-cheng  YANG Guang-ying
Affiliation:ZHANG Shi-qing1,2,ZHAO Zhi-jin1,LEI Bi-cheng2,YANG Guang-ying2(1.School of Telecommunication Engineering,Hangzhou Dianzi University,Hangzhou 310018,China,2.School of Physics , Electronic Engineering,Taizhou College,Linhai 317000,China)
Abstract:Based on extracting fundamental prosody features from emotional speech,speech voice quality features are extracted to improve emotion recognition accuracies.Utilizing support vector machines,we recognize main four speech emotions like anger,happiness,sadness and surprise in Chinese mandarin emotional speech corpus by combining voice quality and prosody features.The experimental results show that,the single prosody features yield an overall accuracy of 86.1%,whereas combining voice quality and prosody featur...
Keywords:prosody features  voice quality features  support vector machines  speech emotion recognition  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号