情感语音的非线性动力学特征

doi:10.3969/j.issn.1001-2400.2016.05.029

西安电子科技大学学报

情感语音的非线性动力学特征

姚慧;孙颖;张雪英

(太原理工大学信息工程学院，山西太原 030024)

收稿日期:2015-06-15 出版日期:2016-10-20 发布日期:2016-12-02
作者简介:姚慧(1991- )，女，太原理工大学硕士研究生，E-mail：5366970@qq.com．
基金资助:
国家自然科学基金资助项目(61371193)；山西省青年科技研究基金资助项目(2013021016-2)；山西省回国留学人员科研资助项目(2013-034)

Research on nonlinear dynamics features of emotional speech

YAO Hui;SUN Ying;ZHANG Xueying

(College of Information Engineering, Taiyuan Univ. of Technology, Taiyuan 030024, China)

Received:2015-06-15 Online:2016-10-20 Published:2016-12-02

摘要/Abstract

摘要：

基于语音发声过程中的混沌特性，提出了非线性动力学模型与情感语音信号处理相结合的方法．提取了该模型下情感语音的非线性特征: 最小延迟时间、关联维数、Kolmogorov熵、最大Lyapunov指数和Hurst指数．设计情感语音识别对比实验以验证非线性特征性能．首先，选用德国柏林语音库和自主录制的TYUT2.0情感语音数据库中的3种情感(高兴、悲伤和愤怒)作为实验数据来源；其次，分别提取非线性特征、韵律特征和梅尔频率倒谱系数特征，采用支持向量机进行了情感识别．结果表明，非线性特征在柏林数据库实验中的识别率高于韵律特征识别率，但是略低于梅尔频率倒谱系数特征识别率，验证了非线性特征是一组区分情感的有效特征; 在TYUT2.0数据库中的识别率均高于韵律特征和梅尔频率倒谱系数特征的识别率，在语料真实度和自然度更高的TYUT2.0数据库中识别结果相对更高，鲁棒性更好．

关键词: 情感语音识别, 混沌特性, 非线性特征, 动力学模型

Abstract:

The application of nonlinear measures based on the chaotic characteristics of emotional speech is proposed. Nonlinear features such as minimum delay time, dimension correlation, Kolmogorov entropy, Lyapunov exponent and Hurst exponent are extracted from the emotional speech signal. The performance of nonlinear features is verified by the comparisons of recognition rates of different features (nonlinear characteristics, prosodic features and MFCC features). First, the Berlin emotional speech database and TYUT2.0 emotional speech database are chosen as the corpus independently, both covering three emotional classifications (anger, happiness and fear). The effectiveness of the nonlinear characteristics is tested on the Support Vector Machine Network. The result shows that the performance of nonlinear features outperforms that of prosodic features on the Berlin emotional speech database and that of prosodic features and MFCC on TYUT2.0 emotional speech database. In addition, nonlinear features have obvious advantage in detecting more natural emotional speech and better robustness.

Key words: emotional speech recognition, chaos theory, nonlinear features, dynamic model

姚慧;孙颖;张雪英. 情感语音的非线性动力学特征[J]. 西安电子科技大学学报, 2016, 43(5): 167-172.

YAO Hui;SUN Ying;ZHANG Xueying. Research on nonlinear dynamics features of emotional speech[J]. Journal of Xidian University, 2016, 43(5): 167-172.

参考文献

［1］韩文静, 李海峰, 阮华斌, 等. 语音情感识别研究进展综述［J］. 软件学报, 2014, 25(1): 37-50.
HAN Wenjing, LI Haifeng, RUAN Huabin, et al. Review on Speech Emotion Recognition ［J］. Journal of Software, 2014, 25(1): 37-50.
［2］ BANBROOK M, MCLAUGHLIN S, MANN I. Speech Characterization and Synthesis by Nonlinear Methods ［J］. IEEE Transactions on Speech and Audio Processing, 1999, 7(1): 1-17.
［3］ LOPEZ-DE-IPINA K, SOLE-CASALS J, EGUIRAUN H, et al. Feature Selection for Spontaneous Speech Analysis to Aid in Alzheimer's Disease Diagnosis: a Fractal Dimension Approach ［J］. Computer Speech and Language, 2015, 30(1): 43-60.
［4］ LOPEZ-DE-IPINA K, ALONSO-HERNANDEZ J B, SOLE-CASALS J, et al. Feature Selection for Automatic Analysis of Emotional Response Based on Nonlinear Speech Modeling Suitable for Diagnosis of Alzheimer's Disease ［J］. Neurocomputing, 2015, 150: 392-401.
［5］李响, 谭南林. 一种应用语音多特征检测驾驶疲劳的方法［J］. 仪器仪表学报, 2013, 34(10): 2231-2237.
LI Xiang, TAN Nanlin. Method of Applying Speech Multi-features to Detect Driver Fatigue ［J］. Chinese Journal of Scientific Instrument, 2013, 34(10): 2231-2237.
［6］ GOMEZ-GARCIA J A, GODINO-LLORENTE J I, CASTELLANOS-DOMINGUEZ G. Non Uniform Embedding Based on Relevance Analysis with Reduced Computational Complexity: Application to the Detection of Pathologies from Biosignal Recording ［J］. Neurocomputing, 2014, 132: 148-158.
［7］ TAKENS F. Detecting Strange Attractors in Turbulence［M］. Berlin: Springer, 1981: 366-381.
［8］ HENRIQUEZ P, ALONSO J B, FERRER M A, et al. Application of Nonlinear Dynamics Characterization to Emotional Speech ［C］//5th International Conference on Nonlinear Speech Processing. Heidelberg: Springer Verlag, 2014: 126-135.
［9］ KANTZ H, SCHREIBER T. Nonlinear Time Series Analysis ［M］. Cambridge: Cambridge University Press, 2004: 30-150.
［10］赵贵兵, 石炎福. 从混沌时间序列同时计算关联维和Kolmogorov熵［J］. 计算物理, 1999, 16(3): 310-315.
ZHAO Guibing, SHI Yanfu. Computing Fractal Dimension and the Kolmogorov Entropy from Chaotic Time Series ［J］. Chinese Journal of Computational Physics, 1999, 16(3): 310-315.
［11］ WOLF A, SWIFT J B, SWINNEY H L, et al. Determining Lyapunov Exponents from a Time Series ［J］. Physica, 1985, 16D(3): 285-317.
［12］ HURST H E, BLACK R P, SIMAIKA Y M. Long-term Storage: an Experimental Study ［J］. Journal of the Royal Statistical Society, 1965, 129(4): 591-593.
［13］ BURKHARDT F, PAESCHKE A, ROLFES M, et al. A Database of German Emotional Speech ［C］//9th European Conference on Speech Communications and Technology. Lisbon: ISCA, 2005: 1517-1520.
［14］赵力, 黄程韦. 实用语音情感识别中的若干关键技术［J］. 数据采集与处理, 2014, 29(2): 157-170.
ZHAO Li, HUANG Chengwei. Key Technologies in Practical Speech Emotion Recognition ［J］. Journal of Data Acquisition and Processing, 2014, 29(2): 157-170.

[1]	万志国,贺王鹏,廖楠楠,窦益华,郭宝龙. 齿根裂纹与齿面剥落故障的振动响应机理研究[J]. 西安电子科技大学学报, 2021, 48(6): 131-137.
[2]	戴显英;郑若川;郭静静;张鹤鸣;郝跃;邵晨峰;吉瑶;杨程. 弛豫锗硅减压化学气相淀积的CFD模型及仿真[J]. J4, 2012, 39(5): 186-191.
[3]	暂时无作者信息. 柔性机械手动力学模型的一个改进[J]. J4, 1997, 24(4): 0-0.

情感语音的非线性动力学特征

Research on nonlinear dynamics features of emotional speech

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

Metrics

本文评价

推荐阅读 0