The track 3.2 round 2 evaluation dataset of ADD 2022

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 (http://addchallenge.cn/add2022) 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. The ADD 2022 dataset is publicly available. This data set is licensed with a CC BY-NC-ND 4.0 license. If you use this dataset, please cite the following paper: Jiangyan Yi, Ruibo Fu, Jianhua Tao, Shuai Nie, Haoxin Ma, Chenglong Wang, Tao Wang, Zhengkun Tian, Ye Bai, Cunhang Fan, Shan Liang, Shiming Wang, Shuai Zhang, Xinrui Yan, Le Xu, Zhengqi Wen, Haizhou Li:ADD 2022: the first Audio Deep Synthesis Detection Challenge. ICASSP 2022: 9216-9220音频深度伪造检测（Audio Deepfake Detection）是一个新兴研究方向，该方向已被纳入ASVspoof 2021挑战赛。然而，现有相关共享任务尚未覆盖诸多真实场景与高难度挑战情境。首届音频深度合成检测挑战赛（Audio Deep Synthesis Detection Challenge，简称ADD）正是为填补这一空白而设立。ADD 2022赛事（官网：http://addchallenge.cn/add2022）共设三个赛道：低质量伪造音频检测（LF）、部分伪造音频检测（PF）以及音频伪造对抗赛（FG）。其中LF赛道聚焦于处理带有各类真实环境噪声的真实语音片段与全伪造语音片段；PF赛道旨在区分部分伪造音频与真实音频；FG赛道为对抗赛事，包含两项子任务：音频生成任务与音频伪造检测任务。本文详细介绍了本次挑战赛的数据集、评估指标与评测流程，并汇报了反映音频深度伪造检测任务当前最新进展的核心研究成果。 ADD 2022数据集已公开可获取。 本数据集采用CC BY-NC-ND 4.0协议进行授权。 若您使用本数据集，请引用如下论文： 姜岩易、傅瑞波、陶建华、聂帅、马浩鑫、王成龙、王涛、田正坤、白烨、范存航、梁珊、王时明、张帅、颜鑫瑞、徐乐、温正琪、李海洲：ADD 2022：首届音频深度合成检测挑战赛，发表于ICASSP 2022，页码9216-9220