The rapid development of voice synthesis and imitation technology poses new challenges to the security of voice-based authentication systems. This study introduces VocaLock, an intelligent access control system that integrates Automatic Speaker Verification (ASV) and Presentation Attack Detection (PAD) to ensure secure voice-based user authentication. The integration of ASV using ECAPA-TDNN and PAD using LCNN produces two layers of security: speaker identity validation and voice authenticity. The system is implemented on an ESP32 microcontroller connected to an IoT-based access control indicator and a cloud server running the verification model. Test results show that VocaLock has high verification performance and is able to resist replay and spoofing attacks with a good level of accuracy. This study contributes to the development of a lightweight, practical, and secure voice biometric framework for application in smart office environments, laboratories, and other confined areas.
Copyrights © 2026