External vs. Internal SVM-RFE: The SVM-RFE Method Revisited and Applied to Emotion Recognition

Hela Daassi-Gnaba, Yacine Oussar

Abstract


Support Vector Machines (SVM) are well known as a kernel based method mostly applied to classication. SVM-Recursive Feature Elimination (SVM- RFE) is a variable ranking and selection method dedicated to the design of SVM based classiers. In this paper, we propose to revisit the SVM-RFE method. We study two implementations of this feature selection method that we call External SVM-RFE and Internal SVM-RFE, respectively. The two implementations are ap- plied to rank and select acoustic features extracted from speech to design optimized linear SVM classiers that recognize speaker emotions. To show the efficiency of the External and Internal SVM-RFE methods, an extensive experimental study is presented. The SVM classiers were selected using a validation procedure that ensures strict speaker independence. The results are discussed and compared with those achieved when the features are ranked using the Gram-Schmidt procedure. Overall, the results achieve a recognition rate that exceeds 90%.

Keywords


feature selection; classification; support vector machines (SVM); emotion recognition

Full Text:

PDF

References


BÄNZIGER T., SCHERER K.R. The role of intonation in emotional expression. Speech Communication. 2005, 46(3-4), pp. 252-267, doi: http://dx.doi.org/10.1016/j.specom.2005.02.016">doi:10.1016/j.specom.2005.02.016" target="_blank">10.1016/j.specom.2005.02.016" target="_blank">http://dx.doi.org/10.1016/j.specom.2005.02.016">doi:10.1016/j.specom.2005.02.016.

BÁRTŮ M. Speech disorder analysis using Matching Pursuit and Kohonen Self-Organizing Maps. Neural Network World. 2012, 22(6), pp. 519-533, doi: 10.14311/NNW.2012.22.032.

Berlin Emotional Speech Database. Available from: http://www.expressive-speech.net/http://www.expressive-speech.net/">http://www.expressive-speech.net/.

BOERSMA P., WEENINK D. Praat: doing phonetics by computer 5.0.32 [software]. 2008-08-12 [accessed 2009-07-15]. Available from: http://www.praat.org/.

BOSER B.E., GUYON I.M., VAPNIK V.N. A training algorithm for optimal margin classifiers. In: Proceedings of the 5th Annual Workshop on Computational Learning Theory (COLT), Pittsburgh, PA, USA. NY, USA: ACM, 1992, pp. 144-152, doi: 10.1145/130385.130401.

CASALE S., et al. Speech emotion classification using machine learning algorithms. In: Proceedings of the IEEE International Conference on Semantic Computing, Santa Clara, CA, USA. MA, USA: IEEE, 2008, pp. 158-165, doi: 10.1109/ICSC.2008.43.

CHEN S., BILLINGS S.A., Luo W. Orthogonal least squares methods and their application to non-linear system identification. International Journal of Control. 1989, 50(5), pp. 1873-1896, doi: 10.1080/00207178908953472.

Corpus LDC. Available from: http://www.ldc.upenn.edu.

CRISTIANINI N., SHAWE-TAYLOR J. Support Vector Machines and other Kernel-based Learning Methods. Cambridge: Cambridge University Press, 2000.

DAASSI-GNABA H., LOPEZ KRAHE J. Universal combined system: speech recognition, emotion recognition and talking head for deaf and hard of hearing people. In: Proceedings of the 10th Association for the Advancement of Assistive Technology in Europe (AAATE), Florence, Italy. Netherlands: IOS Press, 2009, pp. 503-508, doi: 10.3233/978-1-60750-042-1-503.

DAASSI-GNABA H., OUSSAR Y. Enhanced emotion recognition by feature selection to animate a talking head. In: Proceedings of the 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), Bruges, Belgium. Louvain-La-Neuve, Belgium: i6doc.com, 2012, pp. 513-518.

EL AYADI M., KAMEL M.S., KARRAY F. Survey on speech emotion recognition: features, classification schemes, and databases. Pattern Recognition. 2011, 44(3), pp. 572-587, doi: 10.1016/j.patcog.2010.09.020.

GUYON I., ELISSEEFF A. An Introduction to variable and feature selection. Journal of Machine Learning Research. 2003, 3, pp. 1157-1182.

GUYON I., et al. Gene selection for cancer classification using support vector machines. Machine Learning. 2002, 46(1-3), pp. 389-422, doi: 10.1023/A:1012487302797.

HASTIE T., TIBSHIRANI R., FRIEDMAN J. The Elements of Statistical Learning. New-York: Springer, 2009.

HO Y.C., KASHYAP R.L. An algorithm for linear inequalities and its applications. IEEE Transactions on Electron Computer. 1965, 14(5), pp. 683-688, doi: 10.1109/PGEC.1965.264207.

INGALE A.B., CHAUDHARI D.S. Speech emotion recognition. International Journal of Soft Computing and Engineering (IJSCE). 2012, 2(1), pp. 235-238.

KWON O.W., et al. Emotion recognition by speech signals. In: Proceedings of the 8th European Conference on Speech Communication and Technology (EUROSPEECH), Geneva, Switzerland. Bonn, Germany: Wolfgang Hess, 2003, pp. 125-128.

PUDIL P., NOVOVIČOVÁ J., KITTLER J. Floating search methods in feature selection. Pattern Recognition Letters. 1994, 15(11), pp. 1119-1125, doi: 10.1016/0167-8655(94)90127-9.

RABINER L.R., SCHAFER R.W. Introduction to Digital Speech Processing. Hanover: Now Publishers Inc, 2007, doi: 10.1561/2000000001.

SCHULLER B., et al. Speaker independent emotion recognition by early fusion of acoustic and linguistic features within ensembles. In: Proceedings of the 6th Annual Conference of the International Speech Communication Association (ISCA), INTERSPEECH, Lisboa, Portugal. Bonn, Germany: Wolfgang Hess, 2005, pp. 805-809.

SCHULLER B., STEIDL S., BATLINER A. The INTERSPEECH 2009 emotion challenge. In: Proceedings of the 10th Annual Conference of the International Speech Communication Association (ISCA), INTERSPEECH, Brighton, United Kingdom. Bonn, Germany:Wolfgang Hess, 2009, pp. 312-315.

SCHULLER B., et al. The INTERSPEECH 2010 paralinguistic challenge. In: Proceedings of the 11th Annual Conference of the International Speech Communication Association (ISCA), INTERSPEECH, Makuhari, Chiba, Japan. Bonn, Germany: Wolfgang Hess, 2010, pp. 2794-2797.

SCHULLER B., et al. AVEC 2011–The first international audio/visual emotion challenge. In: Proceedings of the 4th International Conference Affective Computing and Intelligent Interaction (ACII), Memphis, Tennessee, USA. Berlin, Germany: Springer, 2011, pp. 415-424.

SCHULLER B., et al. AVEC 2012–The continuous audio/visual emotion challenge. In: Proceedings of the 14th ACM international conference on Multimodal interaction, Santa Monica, California, USA. New York ACM, 2012, pp. 449-456.

SCHULLER B., BATLINER A. Computational Paralinguistics: Emotion, Affect and Personality in Speech and Language Processing. Chichester: Wiley, 2014.

The Spider 1.71 [software]. 2006-07-26 [accessed 2007-10-01]. Available from: http://people.kyb.tuebingen.mpg.de/spider/http://people.kyb.tuebingen.mpg.de/spider/">http://people.kyb.tuebingen.mpg.de/spider/.

TAHON M., DELABORDE A., DEVILLERS L. Real-life emotion detection from speech in human-robot interaction: experiments across diverse corpora with child and adult voices. In: Proceedings of the 12th Annual Conference of the International Speech Communication Association (ISCA), INTERSPEECH, Florence, Italy. Bonn, Germany: Wolfgang Hess, 2011, pp. 3121-3124.

TUCKOVA J., SEBESTA V. Prosody optimisation of a Czech language synthesizer. Neural Network World. 2008, 18(4), pp. 291-308.

YANG N., et al. Speech-based emotion classification using multiclass SVM with hybrid kernel and thresholding fusion. In: Proceedings of the 4th IEEE Workshop on Spoken Language Technology (SLT), Miami, Florida, USA. Miami IEEE, 2012, pp. 455-460, doi: 10.1109/SLT.2012.6424267.




DOI: http://dx.doi.org/10.14311/NNW.2015.25.004

Refbacks

  • There are currently no refbacks.


Should you encounter an error (non-functional link, missing or misleading information, application crash), please let us know at nnw.ojs@fd.cvut.cz.
Please, do not use the above address for non-OJS-related queries (manuscript status, etc.).
For your convenience we maintain a list of frequently asked questions here. General queries to items not covered by this FAQ shall be directed to the journal editoral office at nnw@fd.cvut.cz.