Title :
Speech recognition experiments with guided propagation
Author :
Escande, P. ; Béroule, D. ; Blanchet, P.
Author_Institution :
LIMSI-CNRS, Orsay, France
Abstract :
The authors discuss some features of the associative memory which supports the processing principle called guided propagation, including its requirement for a specific representation of input patterns. The way this approach is currently applied to speech processing is then described. In comparison with a classical speech recognizer, a software simulation of the proposed system is shown to be better at identifying superimposed speech signals
Keywords :
content-addressable storage; neural nets; speech recognition; associative memory; guided propagation; input patterns; software simulation; speech processing; speech recognition; superimposed speech signals; Associative memory; Biological information theory; Biological system modeling; Computer architecture; Delay; Man machine systems; Psychology; Signal processing; Speech processing; Speech recognition;
Conference_Titel :
Neural Networks, 1991. 1991 IEEE International Joint Conference on
Print_ISBN :
0-7803-0227-3
DOI :
10.1109/IJCNN.1991.170492
