Implementation of a word-based speaker authentication system using Altera DE2-115 FPGA board implemented using Verilog HDL

Closed

Harold Reg L. Aviso, James Son G. Pada, Edwin Jude E. Reyes, Percival J. Forcadilla

2016 8th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management, HNICEM 2015 Conference paper Cited by 2 Quartile

Abstract

the need for better and more reliable security measures is prevalent today especially as the commonly existing security measures are getting less reliable. The study was conducted to create a solution to the said problem using the very versatile and relatively inexpensive Field-Programmable Gate Array (FPGA) board. The system incorporates the unique frequency content of every individual as a more reliable and difficult to replicate, replacing the conventional security keys and passwords. The user's frequency content or voice model was extracted using signal processing in FPGA and stored in the system as the basis for comparison for the different speech signals. Input speech signals were matched to the stored voice model and the authentication system shows a locked or unlocked status through an on-board LEDs. LEDs which lights up indicates unlocked and doesn't otherwise. The system was successfully implemented in Altera DE2-115 board, allowing the correct user in unlocking the system while the other incorrect users can't, with a best system identification accuracy of 86%. © 2015 IEEE.

Affiliations

Microelectronics and IT Laboratory, DEEE - University of San Carlos, Philippines