CONCENTRATION INFORMATION IN TIME: ANALOG NEURAL NETWORKS WITH APPLICATIONS TO SPEECH RECOGNITION PROBLEMS.

D. W. Tank, J. J. Hopfield

Research output: Contribution to conferencePaper

14 Scopus citations

Abstract

An electronic implementation of a neural network for a real-time speech recognition problem is described. The circuit is completely analog with the outputs of the circuit driven by the speech input. The computation is totally asynchronous; there is no system clock. The time-dependent synaptic connections and the resulting dynamics of the circuit can be considered as an organizer of information over several time scales.

Original languageEnglish (US)
StatePublished - Dec 1 1987
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Fingerprint Dive into the research topics of 'CONCENTRATION INFORMATION IN TIME: ANALOG NEURAL NETWORKS WITH APPLICATIONS TO SPEECH RECOGNITION PROBLEMS.'. Together they form a unique fingerprint.

  • Cite this