ADD 2022: the First Audio Deep Synthesis Detection Challenge
This work provides a new benchmark for researchers and practitioners in audio deepfake detection, addressing incremental improvements by covering more realistic scenarios.
The paper introduced the first Audio Deep synthesis Detection challenge (ADD 2022) to address gaps in real-life and challenging scenarios for audio deepfake detection, including three tracks: low-quality fake audio detection, partially fake audio detection, and an audio fake game, and reported major findings reflecting recent advances in the field.
Audio deepfake detection is an emerging topic, which was included in the ASVspoof 2021. However, the recent shared tasks have not covered many real-life and challenging scenarios. The first Audio Deep synthesis Detection challenge (ADD) was motivated to fill in the gap. The ADD 2022 includes three tracks: low-quality fake audio detection (LF), partially fake audio detection (PF) and audio fake game (FG). The LF track focuses on dealing with bona fide and fully fake utterances with various real-world noises etc. The PF track aims to distinguish the partially fake audio from the real. The FG track is a rivalry game, which includes two tasks: an audio generation task and an audio fake detection task. In this paper, we describe the datasets, evaluation metrics, and protocols. We also report major findings that reflect the recent advances in audio deepfake detection tasks.