Title :
Unstable connectionist networks in speech recognition
Author :
Rohwer, Richard ; Renals, Stephen ; Terry, Mark
Author_Institution :
Centre for Speech Technol. Res., Edinburgh Univ., UK
Abstract :
Connectionist networks evolve in time according to a prescribed rule. Typically, they are designed to be stable so that their temporal activity ceases after a short transient period. However, meaningful patterns in speech have a temporal component: therefore it seems natural to attempt to map the temporality of speech patterns onto the temporality of an unstable network. The authors have begun some exploratory experiments to train networks to recognise temporal patterns. They have designed fully connected networks that are trained to emulate and classify sequences by regarding each temporal state of a network as a layer in a feedforward network. Training is then performed by a variant of the back-propagation algorithm. They have conducted initial experiments using the output of a peripheral auditory model
Keywords :
neural nets; speech recognition; back-propagation algorithm; feedforward network; fully connected networks; networks training; peripheral auditory model; speech patterns; speech recognition; temporal component; temporal patterns recognition; unstable connectionist networks; Character recognition; Clamps; Feedforward systems; Intelligent networks; Machine learning; Pattern recognition; Simulated annealing; Speech recognition;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1988. ICASSP-88., 1988 International Conference on
Conference_Location :
New York, NY
DOI :
10.1109/ICASSP.1988.196609