Advertisement

Exploring Speech Features for Classifying Emotions along Valence Dimension

  • Shashidhar G. Koolagudi
  • K. Sreenivasa Rao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5909)

Abstract

Naturalness of human speech is mainly because of the embedded emotions. Today’s speech systems lack the component of emotion processing within them. In this work, classification of emotions from the speech data is attempted. Here we have made an effort to search, emotion specific information from spectral features. Mel frequency cepstral coefficients are used as speech features. Telugu simulated emotion speech corpus (IITKGP-SESC) is used as a data source. The database contains 8 emotions. The experiments are conducted for studying the influence of speaker, gender and language related information on emotion classification. Gaussian mixture models are use to capture the emotion specific information by modeling the distribution. An average emotion detection rate of around 65% and 80% are achieved for gender independent and dependent cases respectively.

Keywords

Emotion Emotion recognition Gausssian mixture models Telugu emotional speech database Prosody Spectral features Valence 

References

  1. 1.
    Lee, C.M., Narayanan, S.S.: Toward detecting emotions in spoken dialogs. IEEE Trans. Speech and Audio Processing 13, 293–303 (2005)CrossRefGoogle Scholar
  2. 2.
    Jin, X., Wang, Z.: An Emotion Space Model for Recognition of Emotions in Spoken Chinese. In: Tao, J., Tan, T., Picard, R.W. (eds.) ACII 2005. LNCS, vol. 3784, pp. 397–402. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  3. 3.
    Rao, K.S., Yegnanarayana, B.: Intonation modeling for indian languages. CSL (2008)Google Scholar
  4. 4.
    Yildirim, S., Bulut, M., Lee, C.M., Kazemzadeh, A., Busso, C., Deng., Z., Lee, S., Narayanan, S.: An acoustic study of emotions expressed in speech. In: Int’l Conf. on Spoken Language Processing (ICSLP 2004), Jeju island, Korean (October 2004)Google Scholar
  5. 5.
    Ververidis, D., Kotropoulos, C., Pitas, I.: Automatic emotional speech classifcation. In: ICASSP 2004, pp. I593–I596. IEEE, Los Alamitos (2004)Google Scholar
  6. 6.
    Burkhardt, F., Sendlmeier, W.F.: Verification of acousical correlates of emotional speech using formant-synthesis. In: ITRW on Speech and Emotion, Newcastle, Northern Ireland, UK, September 5-7, pp. 151–156 (2000)Google Scholar
  7. 7.
    Oudeyer, P.-Y.: The production and recognition of emotions in speech: features and algorithms. International Journal of Human Computer Studies 59, 157–183 (2003)CrossRefGoogle Scholar
  8. 8.
    Koolagudi, S.G., Maity, S., Kumar, V.A., Chakrabarti, S., Rao, K.S.: IITKGP- SESC: Speech Database for Emotion Analysis. In: Communications in Computer and Information Science, JIIT University, Noida, India, August 17-19. Springer, Heidelberg (2009)Google Scholar
  9. 9.
    Tato, R., Santos, R., Pardo, R.K.J.: Emotional space improves emotion recognition. In: 7th International Conference on Spoken Language Processing, Denver, Colorado, USA, September 16-20 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Shashidhar G. Koolagudi
    • 1
  • K. Sreenivasa Rao
    • 1
  1. 1.School of Information TechnologyIndian Institute of Technology KharagpurKharagpurIndia

Personalised recommendations