MFRASTA: Voice biometric feature using integration of MFCC and RASTA-PLP

Pervasive computing is the area which is growing very rapidly in today’s world. The main concept used in pervasive computing is the advancement of technology and equipment’s computing is going smaller and gaining more power. In pervasive computing environments, users can get services anytime and anywhere but the ubiquity and mobility of the environments bring new security challenges. The user and the service provider do not know each other in advance; they should mutually authenticate each other. The service provider prefers to identify the user based on his voice while the user tends to stay anonymous. In the latest advancements on security, Voice biometric feature is one among others for identification purpose. This paper describes an integrated feature extraction process for user identification. This integrated technique is based on accessing signals of voice by using auditory system of human using a combination of MFCC and RASTA-PLP. A parameter is used to define the performance of the system named as recognition rate. Finally, a comparison is done against the individual techniques like MFCC, PLP, LPC and it shows that integrated method outperforms the individual methods in given environment.