Title :
Maximum mutual information estimation of HMM parameters for continuous speech recognition using the N-best algorithm
Author_Institution :
BBN Syst. & Technol. Corp., Cambridge, MA, USA
Abstract :
An application of discriminative training methods, maximum mutual information (MMI) training, to large-vocabulary continuous speech recognition is described. An algorithm is developed for efficient MMI estimation of HMM parameters, including exponential codebook coefficients, which cannot be estimated using maximum likelihood (ML) methods. Continuous speech recognition performance of the BYBLOS system on the DARPA 1000-word resource management speech corpus is presented
Keywords :
Markov processes; learning systems; parameter estimation; probability; speech recognition; BYBLOS system; DARPA; continuous speech recognition; discriminative training methods; exponential codebook coefficients; hidden Markov model; learning systems; maximum mutual information training; parameter estimation; resource management speech corpus; Hidden Markov models; Management training; Maximum likelihood decoding; Maximum likelihood estimation; Mutual information; Natural languages; Parameter estimation; Resource management; Speech recognition; Vocabulary;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location :
Albuquerque, NM
DOI :
10.1109/ICASSP.1990.115863
