An FPGA implementation of the local global graph-based voice biometric authentication scheme
Raghudeep Kannavara, George S. Bebis, Nikolaos Bourbakis · 2009
In this work, we present a novel method for authentication based on voice biometrics. Local graphs of the pre-processed speech signal are first calculated by feature extraction and combined to form a global graph that is stored in a database for the purpose of authentication. The distance between the global graphs of the presented test speech signal and the stored reference speech signal is then calculated and based on a threshold, speaker authenticity is established.