Skip to main navigation Skip to search Skip to main content

Speaker recognition system with good generalization properties

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

9 Citations (Scopus)

Abstract

This paper presents speaker recognition system possessing very good generalization properties. Relatively low equal error rate for speaker verification and high identification rate for identification are achieved for very short training and testing sequences. This behaviour is achieved for the kernel modification of a classic Ho-Kashyap linear classifier. Achieved results for the new approach are compared with results for the classic GMM and VQ techniques. Speech of a moderately good quality from the Polish speech corpus was used for development of recognition system.

Original languageEnglish
Title of host publicationInternational Conference on Multimedia Computing and Systems -Proceedings
PublisherIEEE Computer Society
Pages206-210
Number of pages5
ISBN (Electronic)9781479938247
DOIs
Publication statusPublished - 25 Sept 2014
Event2014 International Conference on Multimedia Computing and Systems, ICMCS 2014 - Marrakech, Morocco
Duration: 14 Apr 201416 Apr 2014

Publication series

NameInternational Conference on Multimedia Computing and Systems -Proceedings

Conference

Conference2014 International Conference on Multimedia Computing and Systems, ICMCS 2014
Country/TerritoryMorocco
CityMarrakech
Period14/04/1416/04/14

Keywords

  • kernel matrix
  • pattern recognition
  • speaker recognition
  • speaker verification

ASJC Scopus subject areas

  • Media Technology
  • Computer Science Applications
  • Computer Networks and Communications
  • Theoretical Computer Science

Fingerprint

Dive into the research topics of 'Speaker recognition system with good generalization properties'. Together they form a unique fingerprint.

Cite this