Title :
Competitive training: a connectionist approach to the discriminative training of hidden Markov models (speech recognition)
Author_Institution :
Dept. of Eng., Cambridge Univ., UK
Abstract :
Presents hidden Markov models (HMMs) within a connectionist framework and shows how error back propagation can be used to discriminatively train HMM parameters. The relationship between this competitive training approach and conventional Baum-Welch re-estimation is explored and experimental results presented for its application in ergodic HMM architectures.<>
Keywords :
Markov processes; speech recognition; Baum-Welch estimation; HMM parameters; competitive training; connectionist approach; discriminative training; ergodic architectures; hidden Markov models; speech recognition;
Journal_Title :
Communications, Speech and Vision, IEE Proceedings I
