Discriminative training based on an integrated view of MPE and MMI in margin and error space.

Erik McDermott Shinji Watanabe Atsushi Nakamura

Published in: ICASSP (2010)

Keyphrases

discriminative training
automatic speech recognition
hidden markov models
maximum likelihood
low dimensional
feature selection
bayesian networks
support vector
co occurrence
posterior probability
generalization error
maximum margin
probabilistic interpretation