Abstract: “Speaker recognition is an emerging and very important technique in this new era of human-machine interaction. It has two main tasks: speaker identification and speaker verification. In the past various models have been proposed for the identification of speakers with the help of statistical techniques like Hidden Markov Model, Gaussian Mixture Model. As the Artificial Neural Networks (ANNs) are the universal classifiers. The present research proposes a novel paradigm which utilizes the strong pattern matching capability of ANNs for identification of speakers. Here ten speech samples are collected from 40 different Mel Frequency Cepstral Coefficients (MFCC) are extracted for all the speakers and these coefficients are used to train ANN and then test signals are validated and verified for ANN as well as for Fuzzy Logic. The results of identification are very encouraging.”

Keywords: Speaker Identification, Artificial Neural Network, Fuzzy Logic, MFCC, Hidden Markov Model, Discrete Fourier Transform, Discrete Cosine Transform