Speaker recognition based on the combination of GMM and SVDD

Score-level combination of subsystems can yield significant performance gains over individual subsystems in speaker recognition. A novel speaker verification method based on support vector data description (SVDD) is proposed to remedy the defect of Gaussian mixture model (GMM) to some extent, and then using the theory of multiple classifier systems (MCS),a new speaker recognition system based on the combination of GMM and SVDD is proposed. Experiments on TIMIT speech database show that the GMM-SVDD model fully utilizes the complementarities of GMM and SVDD to improve the performance obviously in speaker verification, closed-set speaker identification and speaker recognition. Streszczenie. Zaproponowano nowa metode rozpoznawania glosu bazującą na systemie SVDD jako alternatywe dla modelu GMM. Nastepnie wykorzystując teorie wielokrotnego systemu klasyfikacji MCS zaproponowano wykorzystanie polączenia systemow GMM i SVDD. Eksperymenty potwierdzily ze nowy model GMM-SVDD umozliwia ulepszone rozpoznawanie glosu. (Rozpoznawanie glosu bazujące na kombinacji systemow GMM i SVDD)

[1]  Robert P. W. Duin,et al.  Support Vector Data Description , 2004, Machine Learning.

[2]  Zhan Yong-zhao,et al.  Support vector data description discriminant analysis , 2011 .

[3]  Patrick Kenny,et al.  Adaptive score normalization for progressive model adaptation in text independent speaker verification , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[4]  Xiongwei Zhang,et al.  Research on SVDD Applied in Speaker Verification , 2010, J. Digit. Content Technol. its Appl..

[5]  Amit Banerjee,et al.  A support vector method for anomaly detection in hyperspectral imagery , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Néstor Becerra Yoma,et al.  Confidence based multiple classifier fusion in speaker verification , 2008, Pattern Recognit. Lett..

[7]  Douglas A. Reynolds,et al.  Speaker Verification Using Adapted Gaussian Mixture Models , 2000, Digit. Signal Process..

[8]  Douglas E. Sturim,et al.  Speaker adaptive cohort selection for Tnorm in text-independent speaker verification , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[9]  Elizabeth Shriberg,et al.  An anticorrelation kernel for improved system combination in speaker verification , 2008, Odyssey.