Abstract
Quadratic Discrimination Function (QDF) is commonly used in speech emotion recognition, which proceeds on the premise that the input data is normal distribution. In this paper, we propose a transformation to normalize the emotional features, then derivate a Modified QDF (MQDF) to speech emotion recognition. Features based on prosody and voice quality are extracted and Principal Component Analysis Neural Network (PCANN) is used to reduce dimension of the feature vectors. The results show that voice quality features are effective supplement for recognition, and the method in this paper could improve the recognition ratio effectively.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Rosalind W. Picard. Affective Computing. Cambridge, MA, MIT Press, 1997, 47–85.
K. Alter, E. Tank, and S. Kotz. Accentuation and emotions-two different systems. ISCA Workshop (ITRW) on Speech and Emotion, Newcastle, Northern Ireland, 2000, 138–142.
S. Chennoukh, A. Gerrits, G. Miet, and P. Sluijter. Speech enhancement via frequency bandwidth extension using line spectral frequencies. ICASSP’01, Salt Lake City, UT, USA, May 1, 2001, 665–668.
C. E. Osgood, J. G. Suci, and P. H. Tannenbaum. The Measurement of Meaning. Urbana, University of Illinois Press, 1957, 31–76.
Wang Zhiping, Zhao Li, and Zou Cairong. Emotional speech recognition based on modified parameter and distance of statistical model of pitch. Acta Acustica, 31(2006)1, 29–34 (in Chinese). 王治平, 赵力, 邹采荣. 基于基音参数规整及统计分布模型距离的语音情感识别. 声学学报, 31(2006)1, 29–34.
Xie Bo, Chen Ling, Chen Gencai, and Chen Chun. Feature selection for emotion recognition of mandarin speech. Journal of Zhejiang University, 41(2007)11, 1816–1822 (in Chinese). 킻늨, 돂쇫, 돂룹닅, 돂뒿. 웕춨뮰폯틴쟩룐쪶뇰뗄쳘 헷톡퓱벼쫵. 헣붭듳톧톧놨(릤톧 냦), 41(2007)11, 1816–1822.
Christer Gobl and Ailbhe Ni Chasaide. The role of voice quality in communicating emotion, mood and attitude. Speech Communication, 40(2003)1–2, 189–212.
Jin Xuecheng and Wang Zengfu. An emotion space model for recognition of emotions in spoken Chinese. ACII2005, Beijing, 2005, 397–402.
Jin Xuecheng. A study on recognition of emotions in speech. [Ph.D. dissertation], Hefei, University of Science and Technology of China, 2007 (in Chinese). 뷰톧돉. 믹폚폯틴탅뫅뗄쟩룐쪶뇰퇐뺿. [늩쪿톧캻싛 컄], 뫏럊, 훐맺뿆톧벼쫵듳톧, 2007.
P. M. Sakia. The Box-Cox transformation technique: a review. The Statistician, 42(1992)2, 169–178.
J. W. Tukey. The comparative anatomy of transformations. Annals of Mathematical Statistics, 28 (1957)3, 602–632.
Tian Jun-feng and Liu Xian-yue. Data classify model of intrusion detection-PCANN. Microelectronics & Computer, 24(2007)09, 126–133 (in Chinese). 田俊峰, 刘仙跃. 入侵检测数据分类模型-PCANN. 微电子学与计算机, 24(2007)09, 126–133.
Author information
Authors and Affiliations
Corresponding author
Additional information
Supported by the Ministry of Education Fund (No: 20050286001), Ministry of Education “New Century Talents Support Plan” (No:NCET-04-0483) and Doctoral Foundation of Ministry of Education (No:20050286001).
Communication author: Zhao Yan, born in 1978, female, Ph.D. candidate.
About this article
Cite this article
Zhao, Y., Zhao, L., Zou, C. et al. Speech emotion recognition using modified quadratic discrimination function. J. Electron.(China) 25, 840–844 (2008). https://doi.org/10.1007/s11767-008-0041-8
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11767-008-0041-8
Key words
- Speech emotion recognition
- Principal Component Analysis Neural Network (PCANN)
- Modified Quadratic Discrimination Function (MQDF)