ABSTRACT
Speaker recognition is defined as the ability to recognize an individual based on the speech signal. The speech signals are recorded using microphones and the data stored in .wav format. Since the speaker recognition systems are mostly used in security based applications modeling, the distribution in the presence of noise and extracting the features efficiently with moderate or low signals is very much needed. The input speech signal is processed and converted in to machine readable format. The features are extracted using feature extraction algorithm. The identity of the speaker is then established with the matching of those feature vectors with that of the data available in database. Several models are utilized for the effective recognition of the speaker. In this paper using different techniques like MFCC, SDC and PCA, the speech is detected along with gamma distribution. The software used for implementing the above techniques is MATLAB.
