Robust text-independent speaker identification using bispectrum slice

Ozkurt T., Akgul T.

IEEE 12th Signal Processing and Communications Applications Conference, Kusadasi, Turkey, 28 - 30 April 2004, pp.418-421 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu.2004.1338552
  • City: Kusadasi
  • Country: Turkey
  • Page Numbers: pp.418-421
  • Istanbul Technical University Affiliated: No

Abstract

In this paper, we propose to use a bispectrurn slice for the mel-frequency cepstrurn coefficients as robust features, to be used in Gaussian mixture model for text-independent speaker identification. In theory, higher order statistics can suppress additive Gaussian noise and save phase information unlike autocorrelation based (power spectral) methods. Feature extraction is achieved through the mel-frequency filter banks, the cosine transform and the logarithm operation to obtain cepstral coefficients. Performance of our proposed features are then compared with the classical mel-frequency cepstrum coefficients under various noisy test uttarances.