Research Paper - A Hybrid Approach for Voice Recognition Using MFCC, VQ and K-Means Algorithm

The speech or voice of any human being is his/her unique personal characteristics. No two have almost identical voice there are some features which must be present in one voice and missing from another voice. In order to identify this uniqueness a robust and efficient technique is required so that we can accurately identify the genuine voice from the bunch of fake voices. The process of recognizing a voice of a given speech from the group of given speakers is called speaker identification. This paper proposes a new improved and efficient technique for voice recognition system by applying Mel Frequency Cepstral Coefficients (MFCC), Vector Quantization, K-means and Euclidean distance technique.