Publication Details
Detecting Spoofing Attacks Using VGG and SincNet: BUT-Omilia Submission to ASVspoof 2019 Challenge
Stafylakis Themos
ATHANASOPOULOU, G.
Rohdin Johan Andréas, M.Sc., Ph.D. (DCGM)
GKINIS, I.
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
Černocký Jan, prof. Dr. Ing. (DCGM)
detecting. spoofing, challenge, AVSSpoof
In this paper, we present the system description of the joint efforts of Brno
University of Technology (BUT) and Omilia - Conversational Intelligence for the
ASVSpoof2019 Spoofing and Countermeasures Challenge. The primary submission for
Physical access (PA) is a fusion of two VGG networks, trained on single and
two-channels features. For Logical access (LA), our primary system is a fusion of
VGG and the recently introduced SincNet architecture. The results on PA show that
the proposed networks yield very competitive performance in all conditions and
achieved 86 % relative improvement compared to the official baseline. On the
other hand, the results on LA showed that although the proposed architecture and
training strategy performs very well on certain spoofing attacks, it fails to
generalize to certain attacks that are unseen during training.
@inproceedings{BUT159993,
author="ZEINALI, H. and STAFYLAKIS, T. and ATHANASOPOULOU, G. and ROHDIN, J. and GKINIS, I. and BURGET, L. and ČERNOCKÝ, J.",
title="Detecting Spoofing Attacks Using VGG and SincNet: BUT-Omilia Submission to ASVspoof 2019 Challenge",
booktitle="Proceedings of Interspeech",
year="2019",
journal="Proceedings of Interspeech",
volume="2019",
number="9",
pages="1073--1077",
publisher="International Speech Communication Association",
address="Graz",
doi="10.21437/Interspeech.2019-2892",
issn="1990-9772",
url="https://www.isca-speech.org/archive/Interspeech_2019/pdfs/2892.pdf"
}