Continuous Speech Classification Systems for Voice Pathologies Identification

Hugo Cordeiro, Carlos Meneses, José Fonseca

Research output: Chapter in Book/Report/Conference proceedingOther chapter contributionpeer-review

19 Citations (Scopus)

Abstract

Voice pathologies identification using speech processing methods can be used as a preliminary diagnostic. The aim of this study is to compare the performance of sustained vowel /a/ and continuous speech task in identification systems to diagnose voice pathologies. The system recognizes between three classes consisting of two different pathologies sets and healthy subjects. The signals are evaluated using MFCC (Mel Frequency Cepstral Coefficients) as speech signal features, applied to SVM (Support Vector Machines) and GMM (Gaussian Mixture Models) classifiers. For continuous speech, the GMM system reaches 74% accuracy rate while the SVM system obtains 72% accuracy rate. For the sustained vowel /a/, the accuracy achieved by the GMM and the SVM is 66% and 69% respectively, a lower result than with continuous speech.
Original languageEnglish
Title of host publicationTechnological Innovation for Cloud-Based Engineering Systems
Pages217-224
Volume450
ISBN (Electronic)978-3-319-16766-4
DOIs
Publication statusPublished - 28 Mar 2015

Publication series

NameTechnological Innovation for Cloud-Based Engineering Systems
Volume450
ISSN (Print)1868-4238
ISSN (Electronic)1868-422X

Keywords

  • Voice pathologies identification
  • Continuous speech
  • Gaussian mixture models
  • Support vector machines

Fingerprint

Dive into the research topics of 'Continuous Speech Classification Systems for Voice Pathologies Identification'. Together they form a unique fingerprint.

Cite this