全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

采用半定规划多核SVM的语音情感识别

DOI: 10.13190/j.jbupt.2015.增.016, PP. 67-71

Keywords: 语音情感识别,多核支持向量机,半定规划

Full-Text   Cite this paper   Add to My Lib

Abstract:

为提高语音情感识别精度,采用二叉树结构设计多分类器,其中使用半定规划法求解并构造多核支持向量机(SVM)分类模型,并采用均方根误差与最大误差对分类器性能进行衡量.对特征选择之后的参数集合进行了测试,结果表明,采用半定规划多核SVM分类模型的情感识别精度达到88.614%,比单核分类模型的识别精度提高了12.376%,且能有效减少误差积累和降低情感状态之间混淆程度.

References

[1]  张石清, 李乐民, 赵知劲. 人机交互中的语音情感识别研究进展[J]. 电路与系统学报, 2013, 18(2): 440-451. Zhang Shiqing, Li Lemin, Zhao Zhijin. A survey of speech emotion recognition in human computer interaction[J]. Journal of Circuits and System, 2013, 18(2): 440-451.
[2]  Moataz ELAyadi, Mohamed S Kamel, Fakhri Karray. Survey on speech emotion recognition: features, classification schemes, and databases[J]. Pattern Recognition, 2011, 44(3): 572-587.
[3]  Lanckriet Gert R G, Cristianini Nello, Bartlett Peter, et al. Learning the kernel matrix with semi-definite programming[J]. Machine Learning Research, 2004, 5(1): 27-72.
[4]  Yeh ChiYuan, Su WenPin, Lee ShieJue. An efficient multiple-kernel learning for pattern classification[J]. Expert Systems with Applications, 2013, 40(9): 3491-3499.
[5]  Meyer Patrick E, Bontempi Gianluca. On the use of variable complementarity for feature selection in cancer classification[C]//Applications of Evolutionary Computing-EvoWorkshops 2006 Proceedings. Budapest: Springer Verlag, 2006: 91-102.
[6]  Wu Siqing, Falk H Tiago, Chan WaiYip. Automatic speech recognition using modulation spectral features[J]. Speech Communication, 2011, 53(5): 768-785.
[7]  Milton A, Tamil Selvi S. Class-specific multiple classifiers scheme to recognize emotions from speech signals[J]. Computer Speech and Language, 2014, 28(3): 717-742.
[8]  Hassan A, Damper R I. Classification of emotional speech using 3DEC hierarchical classifier[J]. Speech Communication, 2012, 54(7): 903-916.
[9]  Jin Yun,Song Peng, Zheng Wenming, et al. Novel feature fusion method for speech emotion recognition based on multiple kernel learning[J]. Journal of Southeast University, 2013, 29(2): 129-133.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133