This paper explores the applicability of Adaptive Resistance Theory- (ART-) type neural networks for finding and encoding linguistic structures, specifically those corresponding to acoustic patterns in natural speech. We build an interpretation of human perceptual response to acoustic pattern in natural speech, translating this to a neural architecture as a model of acquisition, storage, and classification of acoustic speech patterns.
Phoneme Perception Neural Networks Pattern Recognition
Level of Degree
Department of Linguistics
First Committee Member (Chair)
Second Committee Member
Hjelm, Rex Devon. "A Temporal Fuzzy-ART Neural Network Architecture as a Model of Phoneme Perception." (2011). https://digitalrepository.unm.edu/ling_etds/17