Maximum mutual information training of hidden Markov models with vector linear predictors.

K. K. Chin, Philip C. Woodland

Browse the full Interspeech paper archive.