DEVELOPMENT OF A TEXT DEPENDENT SPEAKER RECOGNITION SYSTEM

Abstract

Providing an identity security is one of the most commonly occurring problems in real life. In everyday transaction, you have to prove that you are you. The ability of recognizing a person based on his voice is known as speaker recognition. A speaker recognizer system comprises of two distinct blocks, a feature extractor and a recognizer. The feature extractor block uses MFCCs and LPC analysis technique, with the help of which we can translate the incoming speech into a feature value. Also, a recognizer block, combines the two techniques (MFCC and LPC) to get an hybrid features for speaker identification/verification system. An experiment was carried out on the threshold from 75% to 95% at 5% differential to know the performance of the MFCCs and LPC identification/verification system. From the experiment carried out, the result shows that LPC outperforms MFCCs and a combination, where MFCCs denied a speaker due to the factors like noise, LPC compensate for it to provide recognition accuracy. The text-dependent speaker recognition system was implemented using Java Programming Language. With its application package for speech technology available in Java ( Java Speech Application Programming Interface (JSAPI) ) digitizes the amplitude of each sampled element onto a digital scale and eventually recognized a speaker. The system is developed for access control into computer systems.