Recent singing voice synthesis and conversion advancements necessitate robust singing voice deepfake detection (SVDD) models. Current SVDD datasets face challenges due to limited controllability, diversity in deepfake methods, and licensing restrictions. Addressing these gaps, we introduce CtrSVDD, a large-scale, diverse collection of bonafide and deepfake singing vocals. These vocals are synthesized using state-of-the-art methods from publicly accessible singing voice datasets. CtrSVDD includes 47.64 hours of bonafide and 260.34 hours of deepfake singing vocals, spanning 14 deepfake methods and involving 164 singer identities. We also present a baseline system with flexible front-end features, evaluated against a structured train/dev/eval split. The experiments show the importance of feature selection and highlight a need for generalization towards deepfake methods that deviate further from training distribution. The CtrSVDD dataset and baselines are publicly accessible.
XWSB: A Blend System Utilizing XLS-R and WavLM with SLS Classifier detection system for SVDD 2024 Challenge
College of Intelligent Systems Science and Engineering, Hubei Minzu University（湖北民族大学智能系统科学与工程学院）；School of Mathematics and Statistics, Hubei Minzu University（湖北民族大学数学与统计学院）
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, Chinese Academy of Sciences（中国科学院人工智能学院）；School of Information and Communication Engineering, Communication University of China（中国传媒大学信息与通信工程学院）；Beijing National Research Center for Information Science and Technology, Tsinghua University（北京信息科学与技术国家研究中心，清华大学）；Department of Automation, Tsinghua University（清华大学自动化系）；School of New Media and Communication, Tianjin University（天津大学新媒体与传播学院）；Anhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University（安徽省多模态认知计算重点实验室，安徽大学计算机科学与技术学院）；Tencent, AI Lab（腾讯，AI实验室）
