Абстрактный

Gammatone Cepstral Coefficient for Speaker Identification

Rahana Fathima, Raseena P E

Digital processing of speech signal and voice recognition algorithm is very important for fast and accurate automatic voice recognition technology. The voice is a signal of infinite information. A direct analysis and synthesizing the complex voice signal is due to too much information contained in the signal. Taking as a basis Mel frequency cepstral coefficients (MFCC) used for speaker identification and audio parameterization, the Gammatone cepstral coefficients (GTCCs) are a biologically inspired modification employing Gammatone filters with equivalent rectangular bandwidth bands. A comparison is done between MFCC and GTCC for speaker identification.Thier performance is evaluated using three machine learning methods neural network (NN) and support vector machine (SVM) and K-nearest neighbor (KNN). According to the results, classification accuracies are significantly higher when employing GTCC in speaker identification than MFCC.