UBM-GMM Driven Discriminative Approach for Speaker Verification
作者： Nicolas Scheffer , Jean-francois Bonastre
DOI: 10.1109/ODYSSEY.2006.248127
关键词:
摘要: In the past few years, discriminative approaches to perform speaker detection have shown good results and an increasing interest. Among these methods, SVM based systems lots of advantages, especially their ability deal with a high dimension feature space. Generative such as UBM-GMM show greatest performance among other in verification tasks. Combination generative is not new idea has been studied several times by mapping whole speech utterance onto fixed length vector. This paper presents straight-forward, cost friendly method combine two use UBM model only drive experiment. We that TFLLR kernel, while closely related reduced form Fisher mapping, implies close standard GMM/UBM system. Moreover, we combination both outperforms taken independently.
archives-ouvertes.fr
researchgate.net 
sci-hub.se 参考文章(11)
Elizabeth Shriberg, Luciana Ferrer, Anand Venkataraman, Sachin S. Kajarekar, SVM Modeling of "SNERF-Grams" for Speaker Recognition conference of the international speech communication association. ,(2004)
Mahesan Niranjan, Nathan Smith, Data-dependent Kernels in SVM classification of speech patterns conference of the international speech communication association. pp. 297- 300 ,(2000)
S. Fine, J. Navratil, R.A. Gopinath, A hybrid GMM/SVM approach to speaker identification international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 417- 420 ,(2001) , 10.1109/ICASSP.2001.940856
V. Wan, W.M. Campbell, Support vector machines for speaker verification and identification Neural Networks for Signal Processing X. Proceedings of the 2000 IEEE Signal Processing Society Workshop (Cat. No.00TH8501). ,vol. 2, pp. 775- 784 ,(2000) , 10.1109/NNSP.2000.890157
V. Wan, S. Renals, Speaker verification using sequence discriminant support vector machines IEEE Transactions on Speech and Audio Processing. ,vol. 13, pp. 203- 210 ,(2005) , 10.1109/TSA.2004.841042
Douglas A. Reynolds, Speaker identification and verification using Gaussian mixture speaker models Speech Communication. ,vol. 17, pp. 91- 108 ,(1995) , 10.1016/0167-6393(95)00009-D
J.-F. Bonastre, F. Wils, S. Meignier, ALIZE, a free toolkit for speaker recognition international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 737- 740 ,(2005) , 10.1109/ICASSP.2005.1415219
Frédéric Bimbot, Jean-François Bonastre, Corinne Fredouille, Guillaume Gravier, Ivan Magrin-Chagnolleau, Sylvain Meignier, Teva Merlin, Javier Ortega-García, Dijana Petrovska-Delacrétaz, Douglas A Reynolds, A Tutorial on Text-Independent Speaker Verification EURASIP Journal on Advances in Signal Processing. ,vol. 2004, pp. 430- 451 ,(2004) , 10.1155/S1110865704310024
W.M. Campbell, J.R. Campbell, D.A. Reynolds, D.A. Jones, T.R. Leek, High-level speaker verification with support vector machines international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 73- 76 ,(2004) , 10.1109/ICASSP.2004.1325925
Tommi Jaakkola, David Haussler, Exploiting Generative Models in Discriminative Classifiers neural information processing systems. ,vol. 11, pp. 487- 493 ,(1998)