Autor: |
Fedila, Meriem, Amrouche, Abderrahmane |
Zdroj: |
2012 24th International Conference on Microelectronics (ICM); 1/ 1/2012, p1-4, 4p |
Abstrakt: |
This paper deals with the text-independent speaker recognition over wireless telephone network. The performed speaker verification system, based on Gaussian Mixture Model (GMM), was designed to use the information extracted directly from the coded parameters embedded in the ITU-T G.722.2 bitstream. Experiments were performed over the ARADIGIT database. The robustness in verification accuracy is also studied. The obtained results show that the use of Immittance Spectral Frequency (ISF) features extracted from ITU-T G.722.2 bitstream would improve significantly the recognition performance compared with the Mel Frequency Cepstral Coefficients (MFCC) features extracted from transcoded speech. [ABSTRACT FROM PUBLISHER] |
Databáze: |
Complementary Index |
Externí odkaz: |
|