Summary: HYBRID NEURAL NETWORK/HIDDEN MARKOV MODEL CONTINUOUS­SPEECH RECOGNITION
Michael Cohen*, Horacio Franco*, Nelson Morgan**,
David Rumelhart***, and Victor Abrash*
5
*
* Speech Research Program, SRI International, Menlo Park, CA 9402
* Intl. Computer Science Inst., 1947 Center Street, Suite 600, Berkeley, CA 94704
*** Stanford University, Dept. of Psychology, Stanford, CA 94305
ABSTRACT
n
M
In this paper we present a hybrid multilayer perceptron (MLP)/hidde
arkov model (HMM) speaker­independent continuous­speech recogni­
b
tion system, in which the advantages of both approaches are combined
y using MLPs to estimate the state­dependent observation probabilities
p
of an HMM. New MLP architectures and training procedures are
resented which allow the modeling of multiple distributions for phonetic
a

Source: Abrash, Victor - Speech Technology & Research Laboratory, SRI International

Collections: Computer Technologies and Information Sciences