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

The Auditory Basis of the Perception of Voicing

Robert I. Damper (1,2), S. Harnad (1,3), M. O. Gore (1,2)

(1) Cognitive Sciences Centre; (2) Image, Speech and Intelligent Systems (ISIS) Research Group, Department of Electronics and Computer Science; (3) Department of Psychology
University of Southampton, Southampton, UK

Voicing is perhaps the most basic articulatory and acoustic distinction in speech, yet we know little about the underlying auditory mechanisms of its perception. Responses of both human and animal listeners to synthetic stop-consonant/vowel stimuli in which voice onset time (VOT) is uniformly varied are known to be 'categorical' but an explanation of this phenomenon remains elusive. A 'composite' computational model - consisting of a biologically-realistic auditory model feeding its patterns of neural firing to an artificial neural network (ANN) - is known to be capable of reproducing listeners' behaviour in classical categorical perception (CP) studies. In this paper, we show that the behaviour of the composite model is robust to changes in the ANN architecture and learning algorithm. A contribution analysis of the learned weights reveals that subtle (sub-)features of auditory nerve activity underlie the voiced/unvoiced distinction.

Full Paper

Bibliographic reference.  Damper, Robert I. / Harnad, S. / Gore, M. O. (1996): "The auditory basis of the perception of voicing", In ABSP-1996, 69-74.