DESIGN AND EVALUATION OF A LOW-COST BIOMETRIC WORK ACTIVITY REGISTRATION SYSTEM.

The article focuses on the design and evaluation of a low-cost biometric work activity registration system that uses voice recognition to comply with Spain’s Royal Decree-Law 8/2019, which mandates daily recording of employees' work hours. The proposed system leverages existing Voice over IP (VoIP) infrastructure and an open-source Private Branch Exchange (PBX) called Asterisk, where employees authenticate by repeating randomly selected passphrases verified through text-independent voice biometric analysis and speech-to-text transcription. Experimental results demonstrate high reliability and accuracy in employee identification with passphrases of five words, achieving a 99.7% accuracy and minimizing fraudulent registrations while maintaining usability. The system also includes an alternative identification method for cases where voice recognition fails and addresses potential future threats from real-time AI voice cloning. This approach offers a cost-effective, scalable solution adaptable to various work environments, including remote and mobile employees.