GMM Weights Adaptation Based on Subspace Approaches for Speaker Verification

From LRDE

Abstract

In this paper, we explored the use of Gaussian Mixture Model (GMM) weights adaptation for speaker verifica- tion. We compared two different subspace weight adap- tation approaches: Subspace Multinomial Model (SMM) and Non-Negative factor Analysis (NFA). Both techniques achieved similar results and seemed to outperform the retraining maximum likelihood (ML) weight adaptation. However, the training process for the NFA approach is substantially faster than the SMM technique. The i-vector fusion between each weight adaptation approach and the classical i-vector yielded slight improvements on the tele- phone part of the NIST 2010 Speaker Recognition Eval- uation dataset.


Bibtex (lrde.bib)

@InProceedings{	  dehak.14.odyssey,
  author	= {Najim Dehak and O. Plchot and M.H. Bahari and L. Burget
		  and H. Van hamme and R\'eda Dehak},
  title		= {{GMM} Weights Adaptation Based on Subspace Approaches for
		  Speaker Verification},
  booktitle	= {Odyssey 2014, The Speaker and Language Recognition
		  Workshop},
  year		= 2014,
  address	= {Joensuu, Finland},
  month		= jun,
  abstract	= {In this paper, we explored the use of Gaussian Mixture
		  Model (GMM) weights adaptation for speaker verifica- tion.
		  We compared two different subspace weight adap- tation
		  approaches: Subspace Multinomial Model (SMM) and
		  Non-Negative factor Analysis (NFA). Both techniques
		  achieved similar results and seemed to outperform the
		  retraining maximum likelihood (ML) weight adaptation.
		  However, the training process for the NFA approach is
		  substantially faster than the SMM technique. The i-vector
		  fusion between each weight adaptation approach and the
		  classical i-vector yielded slight improvements on the tele-
		  phone part of the NIST 2010 Speaker Recognition Eval-
		  uation dataset.},
  pages		= {48--53}
}