Linguistics ETDs

Publication Date

7-1-2011

Abstract

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.

Language

English

Keywords

Phoneme Perception Neural Networks Pattern Recognition

Document Type

Thesis

Degree Name

Linguistics

Level of Degree

Masters

Department Name

Department of Linguistics

First Committee Member (Chair)

Morford, Jill

Second Committee Member

Caudell, Thomas

Comments

Submitted by beezbub@unm.edu (beezbub@unm.edu) on 2011-07-15T18:15:01Z No. of bitstreams: 1 Thesis-UNMfinal.pdf: 10165606 bytes, checksum: e7b29954fa7e9f8dbb1129e521789b1c (MD5), Approved for entry into archive by Doug Weintraub(dwein@unm.edu) on 2011-08-30T18:13:37Z (GMT) No. of bitstreams: 1 Thesis-UNMfinal.pdf: 10165606 bytes, checksum: e7b29954fa7e9f8dbb1129e521789b1c (MD5), Made available in DSpace on 2011-08-30T18:13:37Z (GMT). No. of bitstreams: 1 Thesis-UNMfinal.pdf: 10165606 bytes, checksum: e7b29954fa7e9f8dbb1129e521789b1c (MD5)

Share

COinS