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.
Bibliographic reference. Huang, Jui-Ting / Hasegawa-Johnson, Mark (2010): "Semi-supervised training of Gaussian mixture models by conditional entropy minimization", In INTERSPEECH-2010, 1353-1356.