PAC-MFCC feature with sub-band classifier for robust speaker recognition

In this paper we propose the use of the PAC-MFCC (Phase Auto Correlation with Mel frequency of Cepstral Coefficients) feature with Sub-band Classifier for robust speaker recognition. Simulation results show that the proposed scheme outperforms other techniques of the literature for signals degraded by white, factory, babble and pink noise from the Noisex database.

[1]  Kuldip K. Paliwal,et al.  Robust speech recognition in noisy environments based on subband spectral centroid histograms , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[2]  Hervé Bourlard,et al.  Phase autocorrelation (PAC) derived robust speech features , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..