A DCT based nonlinear predictive coding for feature extraction in speech recognition systems

Mahmood Yousefi Azar, Farbod Razzazi

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

6 Citations (Scopus)

Abstract

Speech representation strategies play a key role in automatic speech recognition systems. In this study, a nonlinear procedure has been proposed to overcome the complexities of speech sequence representations. The proposed method may he considered as an extension of nonlinear predictive coding representation procedure in cosine transform domain. The best results belong to classification of nonlinear behaved stop phonemes (i.e. /b/, /d/, /g/) in TIMIT database which show good performance while reducing the computational complexity in comparison to standard NPC.

Original languageEnglish
Title of host publicationCIMSA 2008 - IEEE Conference on Computational Intelligence for Measurement Systems and Applications Proceedings
Pages19-22
Number of pages4
DOIs
Publication statusPublished - 26 Sep 2008
Externally publishedYes
Event2008 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications, IEEE CIMSA 2008 - Istanbul, Turkey
Duration: 14 Jul 200816 Jul 2008

Conference

Conference2008 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications, IEEE CIMSA 2008
CountryTurkey
CityIstanbul
Period14/07/0816/07/08

Keywords

  • Automatic feature extraction
  • Automatic speech recognition
  • Cosine transform
  • Neural network

Fingerprint

Dive into the research topics of 'A DCT based nonlinear predictive coding for feature extraction in speech recognition systems'. Together they form a unique fingerprint.

Cite this