A discriminative training criterion and an associated EM learning algorithm

In this paper we propose a new discriminative training criterion. The criterion is based on a probabilistic interpretation of the minimum classification error (MCE) training and on a modified objective function. A sequential expectation-maximization (EM) based algorithm is derived for optimizing the proposed criterion as an alternative to the well known generalized probabilistic descent (GPD) algorithm. Several variants of the training algorithm are also suggested. The superiority of all variants is shown on a oral/nasal vowel classification task and on a speech/music discrimination task in broadcast news, compared to a maximum likelihood-based system,