Semi-supervised training of Gaussian mixture models by conditional entropy minimization

In this paper, we propose a new semi-supervised training method for Gaussian Mixture Models. We add a conditional entropy minimizer to the maximum mutual information criteria, which enables to incorporate unlabeled data in a discriminative training fashion. The training method is simple but surprisingly effective. The preconditioned conjugate gradient method provides a reasonable convergence rate for parameter update. The phonetic classification experiments on the TIMIT corpus demonstrate significant improvements due to unlabeled data via our training criteria.

• Type: Journal
• Year: 2010
• Where: INTERSPEECH
• Authors: Jui-Ting Huang, Mark Hasegawa-Johnson
• Keywords: Conditional Entropy Minimizer, INTERSPEECH 2010, Semi-supervised Training Method, Signal Processing, Training Method