Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

Discriminative Training for Continuous Speech Recognition

Wolfgang Reichl, Günther Ruske

Institute for Human-Machine-Communication, Munich University of Technology, München, Germany

Discriminative training techniques for Hidden-Markov Models were recently proposed and successfully applied for automatic speech recognition. In this paper a discussion of the Minimum Classification Error and the Maximum Mutual Information objective is presented. An extended reestimation formula is used for the HMM parameter update for both objective functions. The discriminative training methods were utilized in speaker independent phoneme recognition experiments and improved the phoneme recognition rates for both discriminative training techniques.

Full Paper

Bibliographic reference.  Reichl, Wolfgang / Ruske, Günther (1995): "Discriminative training for continuous speech recognition", In EUROSPEECH-1995, 537-540.