Advertisement

Evaluation of Lineal Relation between Shifted Delta Cepstral Features and Prosodic Features in Speaker Verification

  • José R. Calvo
  • Dayana Ribas
  • Rafael Fernández
  • Gabriel Hernández
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5197)

Abstract

Abstract. This paper examines the linear relation between Shifted Delta Cepstral (SDC) features and the dynamic of prosodic features. SDC features were reported to produce superior performance to Δ features in Language Identification and Speaker Recognition. A selection of more correlated SDC features is used in speaker verification to evaluate its robustness to channel/handset mismatch. The experiment reflects superior performance of selected SDC features regarding to Δ features in speaker verification using speech samples from NIST 2001 Ahumada database.

Keywords

speaker verification shifted delta cepstral channel mismatch 

References

  1. 1.
    Furui, S.: Cepstral analysis for automatic speaker verification. IEEE Trans. on Audio Speech and Signal Proc. 29(2), 254–272 (1981)Google Scholar
  2. 2.
    Reynolds, D., Andrews, W., Campbell, J., Navratil, J., Peskin, B., Adami, A., Jin, Q., Klusacek, D., Abramson, J., Mihaescu, R., Godfrey, J., Jones, D., Xiang, B.: The SuperSID Project: Exploiting High-level Information for High-accuracy Speaker Recognition. In: Proceedings of the IEEE ICASSP, vol. 4, pp. 784–787 (2003)Google Scholar
  3. 3.
    Torres-Carrasquillo, P., Singer, E., Kohler, M., Greene, R., Reynolds, D., Deller, J.: Approaches to language identification using Gaussian Mixture Models and shifted delta cepstral features. In: Proceedings of ICSLP 2002, pp. 89–92 (2002)Google Scholar
  4. 4.
    Kinnunen, T., Koh, C., Wang, L., Li, H.: Temporal discrete cosine transform: Towards longer term temporal features for speaker verification. In: Proceedings of ICSLP (2006)Google Scholar
  5. 5.
    Calvo, J., Fernández, R., Hernández, G.: Channel / Handset Mismatch Evaluation in a Biometric Speaker Verification using Shifted Delta Cepstral Features. In: Rueda, L., Mery, D., Kittler, J. (eds.) CIARP 2007. LNCS, vol. 4756, pp. 96–105. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  6. 6.
    Bielefeld, B.: Language identification using shifted delta cepstrum. In: Proceedings Fourteenth Annual Speech Research Symposium (1994)Google Scholar
  7. 7.
    Mary, l., Yegnanarayana, B.: Prosodic features for Speaker Verification. In: Proceedings of Interspeech (2006)Google Scholar
  8. 8.
    Adami, A., Mihaescu, R., Reynolds, D., Godfrey, J.: Modeling prosodic dynamics for Speaker Recognition. In: Proceedings of ICASSP (2003)Google Scholar
  9. 9.
    Soong, F., Rosenberg, A.: On the use of instantaneous and transitional spectral information in speaker recognition. IEEE Trans. on Audio Speech and Signal Proc. 36(6), 871–879 (1988)Google Scholar
  10. 10.
    Ortega-Garcia, J., Gonzalez-Rodriguez, J., Marrero-Aguiar, V.: AHUMADA: A large speech corpus in Spanish for speaker characterization and identification. Speech Comm. 31, 255–264 (2000)CrossRefGoogle Scholar
  11. 11.
    Reynolds, D., Quatieri, T., Dunn, R.: Speaker Verification Using Adapted Gaussian Mixture Models. Digital Signal Proc. 10, 19–41 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • José R. Calvo
    • 1
  • Dayana Ribas
    • 1
  • Rafael Fernández
    • 1
  • Gabriel Hernández
    • 1
  1. 1.Advanced Technologies Application Center, CENATAVCuba

Personalised recommendations