Development of a Text-Dependent Speaker Recognition System

摘要: Speaker recognition is the ability of recognizing a person based on his voice. Many modalities and techniques have been applied to achieve task authentication, ranging from retina scans finger prints. This paper presents TextDependent model for automatic Recognition. Features extractor are Mel-Frequency Cepstrum Coefficients (MFCCs) Linear Predictive Coding (LPC) analysis technique which aids translation incoming speech into feature value. Also, recognizer block, employs two (MFCC LPC) get an hybrid features speaker identification/verification system. An experiment was carried out threshold 75% 95% at 5% differential know performance MFCCs LPC From out, result shows that outperforms combination. The text-dependent system implemented using Java Programming Language (Java Speech Application Interface (JSAPI) ). developed access control computer systems could be used where security considered utmost important.