SPEAKER IDENTIFICATION USING GMM WITH MFCC IN PYTHON

The concept of speaker identification falls into the study of processing the digital signal. The previous systems supported acoustic channel including phonetics which are built for Automatic Speech Recognition. To recognize the user these systems, use pattern matching. Over recent advancements in technology, voice recognition is now considered a very important aspect that provides security to one’s electronic gadgets. This technology has a potential to make our lives more safe and secure. In certain ways, such as biometrics that can authenticate the user on the basis of the features of his/her speech signal. This paper develops the improved approaches by the implementing MFCC algorithm along with GMM techniques to spot a speaker. Efficiency of proposed methods are demonstrated and discussed within the experimental study by real-time and live datasets.