Skip to Main Content
Speaker recognition (SR) technique has proven to be an effective tool in many applications. In this study, SR was used to enhance the security of voice over Internet protocol system. This work focused on authentication issues and user identification detection depending on the voice signature of the incoming call. The authors suggest the use of SR concepts to support user identification process. First, a complete SR system based on linear prediction coefficients (LPC) feature extraction and decision based on artificial neural network was described and built using MATLAB package. Then, MATLAB package was used to develop a prototype version of the modified session initiation protocol (SIP), which is used to manage calls establishment among the network nodes. MATLAB codes were written for both SIP server and client nodes. The modified SIP protocol was implemented on an experimental network which is built to demonstrate the operation of the enhanced authentication procedure.
Date of Publication: April 19 2012