The Journal of the Acoustical Society of Korea (한국음향학회지)

The Acoustical Society of Korea (한국음향학회)
권호 표지

Statistical Extraction of Speech Features Using Independent Component Analysis and Its Application to Speaker Identification

  • 장길진 (한국과학기술원 Department of Computer Science) ;
  • 오영환 (한국과학기술원 Department of Computer Science)
  • Published : 2002.05.01
⟨ Previous Next ⟩

Abstract

We apply independent component analysis (ICA) for extracting an optimal basis to the problem of finding efficient features for representing speech signals of a given speaker The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by adapting the basis functions so that each source component is statistically independent. The learned basis functions are oriented and localized in both space and frequency, bearing a resemblance to Gabor wavelets. These features are speaker dependent characteristics and to assess their efficiency we performed speaker identification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features in that they can obtain a higher speaker identification rate.

Keywords

References

  1. IEEE signal processing magazaine v.9 Robust speaker recognition: a feature-based approach R.J.Mammone;X.Zang;R.P.Ramachandran
  2. Signal Processing v.36 Independent component analysis, A new concept? P.Comon https://doi.org/10.1016/0165-1684(94)90029-9
  3. Neural Computation v.7 no.6 An information maximization approach to blind separation and blind deconvolution A.J.Bell;T.J.Sejnowski
  4. In Proceedings of Eurospeech,(Budapest Hungary) Feature vector transformation using independent component analysis and its application to speaker identification G.J.Jang;S.J.Yun;Yung Hwan
  5. In Proceeding of the Workshop on Automatic Speech Recognition and Understanding,(Keystone, co.,USA) Data-drived non-linear mapping for feature extraction in HMM H.Hermansky;S.Sharma;P.Jain
  6. In Proc. ICASSP,(Istanbul, Turkey) v.3 Speech feature extraction using independent component analysis J.H.Lee;H.Y.Jung;T.W.Lee;S.Y.Lee
  7. Nature v.381 Emergence of simple-cell receptive-field properties by learning a spare code for natural images B.A.Olshausen;D.J.Field https://doi.org/10.1038/381607a0
  8. Neural Computation v.11 no.7 Sparse code shrinkage: denoision of nongaussian data by maximum likelihood estimation A.Hyvaerinen https://doi.org/10.1162/089976699300016214
  9. In International Workshop on Independent Component Analysis (ICA'00), (Helsinki) The generalized Gaussian mixture model using ICA T.W.Lee;M.S.Lewicki
  10. Signal Processing v.24 Blind separation of sources, part Ⅰ: An adaptive algorithm based on neuromimeric architecture C.Jutten;J.Herault https://doi.org/10.1016/0165-1684(91)90079-X
  11. IEEE Trans. on Signal Proc. v.45 no.7 Blind souce separation of mixture of independent sources through a quasi-maximum likelihood approach D.T.Pham;P.Garrat https://doi.org/10.1109/78.599941
  12. In Proc. ICASSP,(Salt Lake City,Utah) The statistical structures of male and female speech signals T.W.Lee;G.J.Jang