Speaker-independent vowel recognition: Comparison of backpropagation and trained classification trees

R. A. Cole, Y. K. Muthusamy, L. Atlas, T. Leen, M. Rudnick

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Experiments comparing the performance of trained classification trees to that of multilayer feedforward networks on speaker-independent vowel recognition, using information in a single spectral slice, are presented. The vowel stimuli were exemplars of 12 monophthongal vowels of American English taken from all phonetic contexts in spoken utterances. The training set consisted of 342 vowel tokens provided by 320 speakers, and the test set consisted of 137 tokens provided by a different 100 speakers. The classification trees and neural classifiers were trained and tested on identical data. In addition, experiments were performed to determine the most effective way to present vowel information for classification. The results show that neural nets trained with back-propagation produce better results than classification trees in all comparable experimental conditions.

Original languageEnglish (US)
Title of host publicationProceedings of the Hawaii International Conference on System Science
EditorsLee W. Hoevel, Bruce D. Shriver, Jay F.Jr. Nunamaker, Ralph H.Jr. Sprague, Velijko Milutinovic
PublisherPubl by Western Periodicals Co
Pages132-141
Number of pages10
ISBN (Print)0818620080
StatePublished - 1990
EventProceedings of the Twenty-Third Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track - Kailua-Kona, HI, USA
Duration: Jan 2 1990Jan 5 1990

Publication series

NameProceedings of the Hawaii International Conference on System Science
Volume1
ISSN (Print)0073-1129

Other

OtherProceedings of the Twenty-Third Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track
CityKailua-Kona, HI, USA
Period1/2/901/5/90

ASJC Scopus subject areas

  • General Computer Science

Fingerprint

Dive into the research topics of 'Speaker-independent vowel recognition: Comparison of backpropagation and trained classification trees'. Together they form a unique fingerprint.

Cite this