Class-Discriminative Weighted Distortion Measure for VQ-based Speaker Identification

  • Tomi Kinnunen
  • Ismo Kärkkäinen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


We consider the distortion measure in vector quantization based speaker identification system. The model of a speaker is a codebook generated from the set of feature vectors from the speakers voice sample. The matching is performed by evaluating the distortions between the unknown speech sample and the models in the speaker database. In this paper, we introduce a weighted distortion measure that takes into account the correlations between the known models in the database. Larger weights are assigned to vectors that have high discriminating power between the speakers and vice versa.


Vector Quantization Speaker Recognition Speaker Verification Speaker Identification Distortion Measure 
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.


  1. 1.
    Deller, J. R. Jr., Hansen, J. H. L., Proakis, J. G.: Discrete-time Processing of Speech Signals. Macmillan Publishing Company, New York, 2000.Google Scholar
  2. 2.
    Fant, G.: Acoustic Theory of Speech Production. The Hague, Mouton, 1960.Google Scholar
  3. 3.
    Finan R. A., Sapeluk A. T., Damper R. I.: ”Impostor cohort selection for score normalization in speaker verification,” Pattern Recognition Letters, 18: 881–888, 1997.CrossRefGoogle Scholar
  4. 4.
    Fränti, P., Kivijärvi, J.: „Randomized local search algorithm for the clustering problem,” Pattern Analysis and Applications, 3(4): 358–369, 2000.CrossRefMathSciNetGoogle Scholar
  5. 5.
    Furui, S.: ”Cepstral analysis technique for automatic speaker verification,” IEEE Transactions on Acoustics, Speech and Signal Processing, 29(2): 254–272, 1981.CrossRefGoogle Scholar
  6. 6.
    Furui, S.: ”Recent advances in speaker recognition,” Pattern Recognition Letters, 18: 859–872, 1997.CrossRefGoogle Scholar
  7. 7.
    Gersho, A., Gray, R. M., Gallager, R.: Vector Quantization and Signal Compression. Kluwer Academic Publishers, 1991.Google Scholar
  8. 8.
    He, J., Liu, L., Palm, G.: ”A discriminative training algorithm for VQ-based speaker identification,” IEEE Transactions on Speech and Audio Processing, 7(3): 353–356, 1999.CrossRefGoogle Scholar
  9. 9.
    Jin, Q., Waibel, A.: „A naive de-lambing method for speaker identification,” Proc. ICSLP 2002, Beijing, China, 2000.Google Scholar
  10. 10.
    Kinnunen, T., Fränti, P.: ”Speaker discriminative weighting method for VQ-based speaker identification,” Proc. 3rd International Conference on Audio-and Video-Based Biometric Person Authentication (AVBPA)): 150–156, Halmstad, Sweden, 2001.Google Scholar
  11. 11.
    Kinnunen, T., Kilpeläinen, T., Fränti P.: ”Comparison of clustering algorithms in speaker identification,” Proc. IASTED Int. Conf. Signal Processing and Communications (SPC): 222–227, Marbella, Spain, 2000.Google Scholar
  12. 12.
    Kohonen T.: Self-Organizing Maps. Springer-Verlag, Heidelberg, 1995.Google Scholar
  13. 13.
    Linde, Y., Buzo, A., Gray, R. M.: ”An algorithm for vector quantizer design,” IEEE Transactions on Communications, 28(1): 84–95, 1980CrossRefGoogle Scholar
  14. 14.
    Nolan, F.: The Phonetic Bases of Speaker Recognition. Cambridge CUP, Cambridge, 1983.Google Scholar
  15. 15.
    Rabiner, L., Juang B.: Fundamentals of Speech Recognition. Prentice Hall, 1993.Google Scholar
  16. 16.
    Soong, F. K., Rosenberg, A. E., Juang, B.-H., Rabiner, L. R.: ”A vector quantization approach to speaker recognition,” AT&T Technical Journal, 66: 14–26, 1987.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Tomi Kinnunen
    • 1
  • Ismo Kärkkäinen
    • 1
  1. 1.Department of Computer ScienceUniversity of JoensuuJOENSUUFinland

Personalised recommendations