EEMD-Based Speaker Emotional Analysis for Speech Signal

Abstract:

Ensemble empirical mode decomposition(EEMD) is a newly developed method aimed at eliminating mode mixing present in the original empirical mode decomposition (EMD). To evaluate the performance of this new method, this paper investigates the effect of two parameters pertinent to EEMD: the emotional envelop and the number of emotional ensemble trials. At the same time, the proposed technique has been utilized for four kinds of emotional(angry、happy、sad and neutral) speech signals, and compute the number of each emotional ensemble trials. We obtain an emotional envelope by transforming the IMFe of emotional speech signals, and obtain a new method of emotion recognition according to different emotional envelop and emotional ensemble trials.

• Periodical: Applied Mechanics and Materials (Volumes 121-126)
• Pages: 815-819
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.121-126.815
• Online since: October 2011
• Authors: Yu Qiang Qin, Xue Ying Zhang
• Keywords: Emotional Ensemble Trials, Emotional Signal Decomposition, Ensemble Empirical Mode Decomposition, Speaker Emotional Analysis

References

[1] Ishi, C., Ishiguro, H., Hagita, N. Analysis of the roles and the dynamics of breathy and whispery voice qualities in dialogue speech[C]. EURASIP J. Audio Speech Music Process., article ID 528193, 2010, 12 pages.

DOI: 10.1186/1687-4722-2010-528193

[2] Thomas F. Quatieri. Discrete-time speech signal processing : principles and practice[M]. Beijing: Publishing House of Electronics Industry，2004：468-488.

[3] Wu, S., Falk, T., Chan, W. -Y. Automatic recognition of speech emotion using long-term spectro-temporal features[C]. In: Proc. Internat. Conf. on Digital Signal Processing, 2009, 1–6.

DOI: 10.1109/icdsp.2009.5201047

[4] Julien Fleureau, Amar Kachenoura, Laurent Albera, Jean-Claude Nunes, Lotfi Senhadji. Multivariate empirical mode decomposition and application to multichannel filtering[C]. Signal Processing, In Press, Corrected Proof, 2011: 789-797.

DOI: 10.1016/j.sigpro.2011.01.018

[5] Xu Guanlei, Wang Xiaotong, Xu Xiaogang. Time-varying frequency-shifting signal-assisted empirical mode decomposition method for AM–FM signals[C]. Mechanical Systems and Signal Processing, Volume 23, Issue 8, November 2009, Pages 2458-2469.

DOI: 10.1016/j.ymssp.2009.06.006

[6] Huang NE, Wu ML, Qu WD. Applications of Hilbert-Huang Transform to Non-stationary Financial Time Series Analysis [J]. Appl. Stochastic Models Bus. Indust (S1524-1904), 2003, 19(3): 245-268.

DOI: 10.1002/asmb.501

[7] GUO Shu-Qing. Data discussion of the EMD method in the practice[J]. Signal Processing, 2010, 26(2): 277-285.

[8] Quek S T, Tua P S, Wang Q. Detecting anomalies in beams and plate based on the Hilbert - Huang transform of real signals [J]. Smart Materials and Structures, 2003, 12: 447- 460.

DOI: 10.1088/0964-1726/12/3/316

[9] ZHONG You-ming, JIN Tao, QIN Shu-ren. New Envelope Algorithm for Hilbert-Huang Transform[J]. Journal of Data Acquisition & Processing, 2005, 20(1): 13-17.

[10] Giannakopoulos, T., Pikrakis, A., Theodoridis, S. A dimensional approach to emotion recognition of speech from movies[C]. In: Proc. Internat. Conf. on Acoustics, Speech and Signal Processing, 2009, 65–68.

DOI: 10.1109/icassp.2009.4959521