TY - JOUR
T1 - Wavelet basis selection for enhanced speech parametrization in speaker verification
AU - Mporas, Iosif
AU - Ganchev, Todor
AU - Siafarikas, Mihalis
AU - Stoyanova, Tsenka
N1 - Todor Ganchev, Mihalis Siafarikas, Iosif Mporas, and Tsenka Stoyanova, 'Wavelet basis selection for enhanced speech parametrization in speaker verification', International Journal of Speech Technology, vol. 17 (1): 27-36, June 2013, doi: https://doi.org/10.1007/s10772-013-9202-8.
Published by Springer US.
PY - 2014/3/31
Y1 - 2014/3/31
N2 - We study the inherent properties of nine wavelet functions and subsequently evaluate their applicability as basis functions in a speech parametrization scheme that is advantageous for speaker verification. Particularly, the inherent properties of nine candidate basis functions are initially analysed and their advantages and disadvantages are discussed. Subsequently, all candidates are employed in a well-proven speech parametrization scheme, and the resulting speech features are computed. Finally, these speech features are evaluated in a common experimental set-up on the speaker verification task. The experimental results, obtained on two well-known speaker recognition databases, show that the Battle-Lemarié wavelet function is the most advantageous one, among all other functions evaluated here, since it leads to the most beneficial speech descriptors. When compared to the baseline Mel-frequency cepstral coefficients (MFCC), a relative reduction of the equal error rate by 4.2 % was observed on the 2001 NIST speaker recognition evaluation database, and by 2.3 % on the Polycost speaker recognition database.
AB - We study the inherent properties of nine wavelet functions and subsequently evaluate their applicability as basis functions in a speech parametrization scheme that is advantageous for speaker verification. Particularly, the inherent properties of nine candidate basis functions are initially analysed and their advantages and disadvantages are discussed. Subsequently, all candidates are employed in a well-proven speech parametrization scheme, and the resulting speech features are computed. Finally, these speech features are evaluated in a common experimental set-up on the speaker verification task. The experimental results, obtained on two well-known speaker recognition databases, show that the Battle-Lemarié wavelet function is the most advantageous one, among all other functions evaluated here, since it leads to the most beneficial speech descriptors. When compared to the baseline Mel-frequency cepstral coefficients (MFCC), a relative reduction of the equal error rate by 4.2 % was observed on the 2001 NIST speaker recognition evaluation database, and by 2.3 % on the Polycost speaker recognition database.
U2 - 10.1007/s10772-013-9202-8
DO - 10.1007/s10772-013-9202-8
M3 - Article
SN - 1381-2416
VL - 17
SP - 27
EP - 36
JO - International Journal of Speech Technology
JF - International Journal of Speech Technology
IS - 1
ER -