Biometric identity verification

Maciej Smiatacz

Biometric identity verification

Abstract

This chapter discusses methods which are capable of protecting automatic speaker verification systems (ASV) from playback attacks. Additionally, it presents a new approach, which uses computer vision techniques, such as the texture feature extraction based on Local Ternary Patterns (LTP), to identify spoofed recordings. We show that in this case training the system with large amounts of spectrogram patches may be difficult, and concentrate on feature engineering issues. The goal is to make the algorithm independent from the contents of the training set as much as possible; we look for the descriptors that would allow the method to detect attacks performed in an environment entirely different from the training one and with the use of the equipment that differs considerably from the devices that captured the training samples. The final form of the new method performs significantly better than the reference Textrogram algorithm.