Speaker identification security improvement by means of speech watermarking

Abstract

This paper presents a security enhanced speaker identification system based on speech signal watermarking. Our proposed system can detect several situations where a playback speech, a synthetically generated speech, or a hacker trying to imitate the speech is fooling the biometric system. It is also suitable for forensic experts, who sometimes have to demonstrate in front of a court that a digital recording has neither been manipulated nor edited. In addition, we demonstrate that this watermark can coexist simultaneously with biometric speaker identification based on Gaussian mixture models (GMM), minimizing the mutual effects.