A Review of Comparatively Study of Different Speaker Recognition Techniques

In this paper, we describe a brief overview of the speaker recognition techniques with their processing steps. Speaker recognition has many problems in feature extraction due to the robustness of the speech with noise. Gamma Tone Filter Bank and Wavelet Packet for the speaker recognition have the best performance over the Hidden Markov Model, Mel Frequency Cepstral Coefficient, Dynamic Time Warping and Layered Recurrent Neural Network. The system performance was measured by recognition rate with various signal-to-noise ratios over -10 to 10 dB.