Voice Based Authentication: A Practical and Privacy-Preserving Solution

Abstract

This project proposes an efficient algorithm to perform privacy-preserving speaker authentication. We consider a scenario where the users have to enroll their voice biometrics with the third-party service providers to access services (i.e., banking). Once the enrolment is done, the users can authenticate themselves to the service provider using their voice instead of passwords. Since the voice is unique for individuals, storing this on the third-party server raises privacy concerns. Hence, in this paper we propose a novel technique using randomization to perform voice authentication, which allows users to enroll and authenticate their voice in the encrypted domain, hence privacy is preserved. The proposed algorithm is validated using the widely used TIMIT speech corpus. Experimental results demonstrate that the proposed algorithm does not degrade the performance compared to existing works.