Advertisement

Global Impostor Selection for DBNs in Multi-session i-Vector Speaker Recognition

  • Omid Ghahabi
  • Javier Hernando
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8854)

Abstract

An effective global impostor selection method is proposed in this paper for discriminative Deep Belief Networks (DBN) in the context of a multi-session i-vector based speaker recognition. The proposed method is an iterative process in which in each iteration the whole impostor i-vector dataset is divided randomly into two subsets. The impostors in one subset which are closer to each impostor in another subset are selected and impostor frequencies are computed. At the end, those impostors with higher frequencies will be the global selected ones. They are then clustered and the centroids are considered as the final impostors for the DBN speaker models. The advantage of the proposed method is that in contrary to other similar approaches, only the background i-vector dataset is employed. The experimental results are performed on the NIST 2014 i-vector challenge dataset and it is shown that the proposed selection method improves the performance of the DBN-based system in terms of minDCF by 7% and the whole system outperforms the baseline in the challenge by more than 22% relative improvement.

Index Terms: Speaker Recognition, Deep Belief Network, Impostor Selection, NIST i-vector challenge.

Keywords

Speaker Recognition Deep Belief Network Speech Utterance Universal Background Model Target Speaker 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dehak, N., Kenny, P., Dehak, R., Dumouchel, P., Ouellet, P.: Front-End Factor Analysis for Speaker Verification. IEEE Transactions on Audio, Speech, and Language Processing 19, 788–798 (2011)CrossRefGoogle Scholar
  2. 2.
    Prince, S., Elder, J.: Probabilistic Linear Discriminant Analysis for Inferences About Identity. In: IEEE 11th International Conference on Computer Vision, ICCV 2007 (2007)Google Scholar
  3. 3.
    Kenny, P.: Bayesian speaker verification with heavy tailed priors. In: IEEE Odyssey Speaker and Language Recognition Workshop (2010)Google Scholar
  4. 4.
    Brummer, N., Villiers, E.: The speaker partitioning problem. In: Proceedings of the Odyssey Speaker and Language Recognition Workshop, Brno, Czech Republic (2010)Google Scholar
  5. 5.
    Liu, G., Hasan, T., Boril, H., Hansen, J.: An investigation on back-end for speaker recognition in multi-session enrollment. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7755–7759 (2013)Google Scholar
  6. 6.
    Larcher, A., Lee, K., Ma, B., Li, H.: Phonetically-constrained PLDA modeling for text-dependent speaker verification with multiple short utterances. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7673–7677 (2013)Google Scholar
  7. 7.
    Larcher, A., Bonastre, J.-F., Fauve, B., Lee, K., Lvy, C., Li, H., Mason, J., Parfait, J.-Y.: ALIZE 3.0 Open Source Toolkit for State-of-the-Art Speaker Recognition. In: Proc. Interspeech, pp. 2768–2771 (2013)Google Scholar
  8. 8.
    The 2013-2014 Speaker Recognition i-vector Machine Learning Challenge (2014)Google Scholar
  9. 9.
    Khoury, E., El Shafey, L., Ferras, M., Marcel, S.: Hierarchical speaker clustering methods for the NIST i-vector Challenge. In: Odyssey: The Speaker and Language Recognition Workshop, pp. 254–259 (2014)Google Scholar
  10. 10.
    Novoselov, S., Pekhovsky, T., Simonchik, K.: STC Speaker Recognition System for the NIST i-Vector Challenge. In: Odyssey: The Speaker and Language Recognition Workshop, pp. 231–240 (2014)Google Scholar
  11. 11.
    Vesnicer, B., Zganec-Gros, J., Dobrisek, S., Struc, V.: Incorporating Duration Information into I-Vector-Based Speaker-Recognition Systems. In: Odyssey: The Speaker and Language Recognition Workshop, pp. 241–248 (2014)Google Scholar
  12. 12.
    Khosravani, A., Homayounpour, M.: Linearly Constrained Minimum Variance for Robust I-vector Based Speaker Recognition. In: Odyssey: The Speaker and Language Recognition Workshop, pp. 249–253 (2014)Google Scholar
  13. 13.
    Ghahabi, O., Hernando, J.: i-Vector Modeling with Deep Belief Networks for Multi-Session Speaker Recognition. In: Odyssey: The Speaker and Language Recognition Workshop, pp. 305–310 (2014)Google Scholar
  14. 14.
    McLaren, M., Baker, B., Vogt, R., Sridharan, S.: Improved SVM speaker verification through data-driven background dataset collection. In: IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2009, pp. 4041–4044 (2009)Google Scholar
  15. 15.
    Liu, G., Suh, J.-W., Hansen, J.: A fast speaker verification with universal background support data selection. In: 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4793–4796 (2012)Google Scholar
  16. 16.
    Ghahabi, O., Hernando, J.: Deep belief networks for i-vector based speaker recognition. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2014)Google Scholar
  17. 17.
    Hinton, G., Osindero, S., Teh, Y.-W.: A Fast Learning Algorithm for Deep Belief Nets. Neural Computation 18, 1527–1554 (2006)CrossRefzbMATHMathSciNetGoogle Scholar
  18. 18.
    Hinton, G., Salakhutdinov, R.: Reducing the Dimensionality of Data with Neural Networks. Science 313, 504–507 (2006)CrossRefzbMATHMathSciNetGoogle Scholar
  19. 19.
    Hinton, G.E.: A Practical Guide to Training Restricted Boltzmann Machines. In: Montavon, G., Orr, G.B., Müller, K.-R. (eds.) NN: Tricks of the Trade, 2nd edn. LNCS, vol. 7700, pp. 599–619. Springer, Heidelberg (2012)Google Scholar
  20. 20.
    Dahl, G., Yu, D., Deng, L., Acero, A.: Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition. IEEE Transactions on Audio, Speech, and Language Processing 20, 30–42 (2012)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Omid Ghahabi
    • 1
  • Javier Hernando
    • 1
  1. 1.TALP Research Center, Department of Signal Theory and CommunicationsUniversitat Politecnica de Catalunya - BarcelonaTechSpain

Personalised recommendations