CAU_KU team's submission to ADD 2022 Challenge task 1: Low-quality fake audio detection through frequency feature masking

2022/02/09 by Kwak, Il-Youp, Choi, Sunmook, Yang, Jonghoon +2
#Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2202.04328

Abstract

This technical report describes Chung-Ang University and Korea University (CAU_KU) team's model participating in the Audio Deep Synthesis Detection (ADD) 2022 Challenge, track 1: Low-quality fake audio detection. For track 1, we propose a frequency feature masking (FFM) augmentation technique to deal with a low-quality audio environment. %detection that spectrogram-based models can be applied. We applied FFM and mixup augmentation on five spectrogram-based deep neural network architectures that performed well for spoofing detection using mel-spectrogram and constant Q transform (CQT) features. Our best submission achieved 23.8% of EER ranked 3rd on track 1.

Related

• Audio Deep Fake Detection System with Neural Stitching for ADD 2022
• ADD 2022: the First Audio Deep Synthesis Detection Challenge
• Audios Don't Lie: Multi-Frequency Channel Attention Mechanism for Audio Deepfake Detection
• ADD 2023: the Second Audio Deepfake Detection Challenge
• Complex-valued neural networks for voice anti-spoofing