Open AccessAutomatic Speaker Recognition Dependency on Both the Shape of Auditory Critical Bands and Speaker Discriminative MFCCs
Author(s) -
Ivan Jokić,
Vlado Delić,
S. Jokić,
Zoran Perić
Publication year - 2015
Publication title -
advances in electrical and computer engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 23
eISSN - 1844-7600
pISSN - 1582-7445
DOI - 10.4316/aece.2015.04004
Subject(s) - speech recognition , discriminative model , speaker recognition , computer science , dependency (uml) , speaker diarisation , critical band , artificial intelligence , pattern recognition (psychology)
Accuracy of an automatic speaker recognition system predominantly depends on speaker models and features that are used. An influence of the shape of auditory critical bands and a contribution of individual components of MFCC-based feature vectors are investigated in the paper and some experimental results are presented and showed their impact on the accuracy of automatic speaker recognition. The speaker-discrimination capability of the MFCCs was experimentally determined by comparing training and test models for the same speaker. The experiments are conducted with three speech databases and showed that 0th and 19th (the last one) MFCCs are non speaker discriminative. The values of MFCCs are determined by the type of applied auditory critical band. The exponential auditory critical bands based on the lower part of exponential function have outperformed the speaker recognition accuracy of other auditory critical bands such as rectangular or triangular shape