Joint MFCC-and-vector quantization based text-independent speaker recognition system

Signal processing front end for extracting the feature set is an important stage in any speaker recognition system. There are many types of features that are derived differently and have good impact on the recognition rate. This paper uses one of the techniques to extract the feature set from a speech signal known as Mel Frequency Cepstrum Coefficients (MFCCs) to represent the signal parametrically for further processing. Speakers provide samples of their voices once in a training session and once in a testing session later. Subsequently, the feature coefficients {MFCCs} are calculated in both phases and the speaker is identified according to the minimum quantization distance which is calculated between the stored features in the training phase and the MFCCs of the speaker who requests to log into the system in the testing phase. The proposed recognition system was designed and implemented using three different algorithms in MATLAB. Simulation and experimental results show that the Joint MFCC-and-vector quantization algorithm achieves better performance compared to the MFCC and FFT algorithms in terms of recognition accuracy and text dependency.