[1]王富,孙林慧,苏敏,等.基于参数寻优决策树SVM 的语音情感识别[J].计算机技术与发展,2018,28(07):63-67.[doi:10.3969/ j. issn.1673-629X.2018.07.014]
　WANG Fu,SUN Lin-hui,SU Min,et al.Speech Emotion Recognition of Decision Tree SVM Based on Parameter Optimization[J].,2018,28(07):63-67.[doi:10.3969/ j. issn.1673-629X.2018.07.014]
点击复制

基于参数寻优决策树SVM 的语音情感识别()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年07期
页码:
63-67
栏目:
智能、算法、系统工程
出版日期:
2018-07-10

文章信息/Info

Title:
Speech Emotion Recognition of Decision Tree SVM Based on Parameter Optimization
文章编号:
1673-629X(2018)07-0063-05
作者:
王富; 孙林慧; 苏敏; 赵城
南京邮电大学 通信与信息工程学院 宽带无线通信与传感网技术教育部重点实验室,江苏 南京 210003
Author(s):
WANG Fu; SUN Lin-hui; SU Min; ZHAO Cheng
Key Lab of Broadband Wireless Communication and Sensor Network Technology of Ministry of Education,School of Communication and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)
关键词:
语音情感识别; 情感混淆度; 决策树 SVM; 遗传算法; 参数寻优
Keywords:
speech emotion recognition; emotional confusion; decision tree SVM; genetic algorithm; parameter optimization
分类号:
TN912.3
DOI:
10.3969/ j. issn.1673-629X.2018.07.014
文献标志码:
A
摘要:
在多种情感的语音情感识别中,由于部分情感状态容易混淆,导致语音情感识别的总体识别率降低;同时,对于不同的训练集,SVM 参数惩罚因子和核函数参数对识别结果也存在一定影响。 为了有效提高语音情感识别系统的识别率,在传统支持向量机(SVM)的基础上,提出了一种基于参数寻优决策树 SVM 的语音情感识别方法。 该方法首先通过计算情感混淆度构建决策树 SVM 框架,然后采用遗传算法对决策树 SVM 中每个 SVM 的惩罚因子和核函数参数进行寻优,最后将参数优化后的决策树 SVM 模型应用于语音情感识别。 在中文情感语音库的实验结果表明,与传统基于 SVM 分类方法的语音情感识别进行对比,该方法可将六种情感的平均识别率提高6.5%。
Abstract:
In the multi-emotion speech emotion recognition,the partial recognition rate is reduced because of the confusion of some emotional states. At the same time,the penalty factors and kernel function parameter also have some influence on the recognition results for different training sets. On the basis of traditional support vector machine (SVM),we propose a decision tree SVM speech emotion recognition algorithm based on parameter optimization for improving the accuracy of speech emotion recognition. In this algorithm,the decision tree SVM framework is firstly established by calculating the confusion degree of emotion. Then the genetic algorithm is used to optimize the penalty factor and kernel function parameters of each SVM in the decision tree SVM. Finally,the decision tree SVM model with optimized parameters is applied to speech emotion recognition. The experiment on the Chinese emotion speech database shows that the pro-
posed method can improve the average recognition rate of 6 emotions by 6. 5%,compared with speech recognition based on traditional SVM classification algorithm.

相似文献/References:

[1]石瑛 胡学钢 方磊.基于决策树的多特征语音情感识别[J].计算机技术与发展,2009,(01):147.
　SHI Ying,HU Xue-gang,FANG Lei.Research of Speech Emotion Recognition Based on Decision Tree and Acoustic Features[J].,2009,(07):147.
[2]王健,韩志艳.基于正交实验设计的语音情感识别参数优化[J].计算机技术与发展,2013,(03):109.
　WANG Jian,HAN Zhi-yan.Parameter Optimization of Speech Emotion Recognition Based on Orthogonal Test Design[J].,2013,(07):109.
[3]林巧民,齐柱柱.基于 HMM 和 ANN 混合模型的语音情感识别研究[J].计算机技术与发展,2018,28(10):74.[doi:10.3969/ j. issn.1673-629X.2018.10.015]
　LIN Qiao-min,QI Zhu-zhu.Research on Speech Emotion Recognition Based on HMM and ANN Mixed Model[J].,2018,28(07):74.[doi:10.3969/ j. issn.1673-629X.2018.10.015]
[4]李 鹤,冉 妮,王 蔚.基于知识图谱的语音情感识别研究分析[J].计算机技术与发展,2020,30(06):135.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 026]
　LI He,RAN Ni,WANG Wei.Research and Analysis of Speech Emotion Recognition Based on Knowledge Map[J].,2020,30(07):135.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 026]
[5]吴俊清,倪建成,魏媛媛.语音情感识别中面向小数据集的 CGRU 方法[J].计算机技术与发展,2020,30(12):77.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 014]
　WU Jun-qing,NI Jian-cheng,WEI Yuan-yuan.CGRU Method for Small Datasets in Speech Emotion Recognition[J].,2020,30(07):77.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 014]

更新日期/Last Update: 2018-08-27