Workshop on the Auditory Basis of Speech Perception Keele University, UK July 15-19, 1996

Speech Recognition Using a Combination of Auditory Models and Conventional Neural Networks

Michael J. Pont, Paul I. J. Keeton, Pramod Palooran

BTSP: Speech and Hearing, Department of Engineering, University of Leicester, Leicester, UK

This paper compares the utility of simulated auditory nerve and dorsal cochlear nucleus representations of sound for the purposes of speech recognition. Under test was the hypothesis that the "higher" up the auditory pathway we go, the "better" the representation we would find. In relation to the initial hypothesis, the results from this study are interesting but inconclusive.

Full Paper

Bibliographic reference.  Pont, Michael J. / Keeton, Paul I. J. / Palooran, Pramod (1996): "Speech recognition using a combination of auditory models and conventional neural networks", In ABSP-1996, 321-324.