Gaussian Selection for Speaker Recognition Using Cumulative Vectors

  • Flavio J. Reyes Díaz
  • José Ramón Calvo de Lara
  • Gabriel Hernández Sierra
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7441)

Abstract

Speaker recognition systems frequently use GMM - MAP method for modeling speakers. This method represents a speaker using a Gaussian mixture. However in this mixture not all the Gaussian components are truly representative of the speaker. In order to remove the model redundancy, this work proposes a Gaussian selection method to achieve a new GMM model only with the more representative Gaussian components. Speaker verification experiments applying the proposal show a similar performance to baseline; however the speaker models have a reduction of 80 % regarding the speaker model used for baseline. The application of this Gaussian selection method in real or embedded speaker verification systems could be very useful for reducing computational and memory cost.

Keywords

speaker verification gaussian components selection cumulative vector 

References

  1. 1.
    Reynolds, D.A.: Speaker identification and verification using Gaussian mixture speaker models. Speech Communication 17(1), 91–108 (1995)CrossRefGoogle Scholar
  2. 2.
    Reynolds, D.A., Quatieri, T., Dunn, R.: Speaker verification using adapted gaussian mixture models. Digital Signal Processing 10(1), 19–41 (2000)CrossRefGoogle Scholar
  3. 3.
    Saeidi, R., Sadegh Mohammadi, H.R., Ganchev, T., Rodman, R.D.: Particle Swarm Optimization for Sorted Adapted Gaussian Mixture Models. IEEE Trans. on Audio, Speech, and Language Processing 17(2), 344–353 (2009)CrossRefGoogle Scholar
  4. 4.
    Auckenthaler, R., Mason, J.: Gaussian selection applied to text independent speaker verification. In: Proceedings of Speaker Odyssey: the Speaker Recognition Workshop, Crete, Greece, pp. 83–88 (2001)Google Scholar
  5. 5.
    Xiang, B., Berger, T.: Efficient text-independent speaker verification with structural Gaussian mixture models and neural network. IEEE Transactions on Speech and Audio Processing, 447–456 (2003)Google Scholar
  6. 6.
    Kinnunen, T., Karpov, E., Franti, P.: Real-time speaker identification and verification. IEEE Transaction on Audio, Speech and Language Processing 14(1), 277–288 (2006)CrossRefGoogle Scholar
  7. 7.
    Mohammadi, H.R.S., Saeidi, R.: Efficient implementation of GMM based speaker verification using sorted Gaussian mixture model. In: Proc. EUSIPCO 2006, Florence, Italy (2006)Google Scholar
  8. 8.
    Saeidi, R., Kinnunen, T., Mohammadi, H.R.S., Rodman, R., Fränti, P.: Joint frame and gaussian selection for text independent speaker verification. In: IEEE Trans. ICASSP 2010, pp. 4530–4533 (2010)Google Scholar
  9. 9.
    Liu, Q., Huang, W., Xu, D., Cai, H., Dai, B.: A fast implementation of factor analysis for speaker verification. In: Interspeech 2010, pp. 1077–1080 (2010)Google Scholar
  10. 10.
    Anguera, X., Bonastre, J.F.: A Novel Speaker Binary Key Derived from Anchor Models. In: Proceedings of Interspeech (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Flavio J. Reyes Díaz
    • 1
  • José Ramón Calvo de Lara
    • 1
  • Gabriel Hernández Sierra
    • 1
  1. 1.Advanced Technologies Application CenterCuba

Personalised recommendations