All existing databases of spoofed speech contain attack data that is spoofed in its entirety. In practice, it is entirely plausible that successful attacks can be mounted with utterances that are only partially spoofed. By definition, partially-spoofed utterances contain a mix of both spoofed and bona fide segments, which will likely degrade the performance of countermeasures trained with entirely spoofed utterances. This hypothesis raises the obvious question: 'Can we detect partially-spoofed audio?' This paper introduces a new database of partially-spoofed data, named PartialSpoof, to help address this question. This new database enables us to investigate and compare the performance of countermeasures on both utterance- and segmental- level labels. Experimental results using the utterance-level labels reveal that the reliability of countermeasures trained to detect fully-spoofed data is found to degrade substantially when tested with partially-spoofed data, whereas training on partially-spoofed data performs reliably in the case of both fully- and partially-spoofed utterances. Additional experiments using segmental-level labels show that spotting injected spoofed segments included in an utterance is a much more challenging task even if the latest countermeasure models are used.
Spoof Diarization: "What Spoofed When" in Partially Spoofed Audio
National Institute of Informatics（信息学研究所）；SOKENDAI（高级研究大学机构）；National Institute of Information and Communications Technology（信息通信技术研究所）；Brno University of Technology（布尔诺理工大学）；EURECOM（欧洲通信与网络研究中心）
ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
EURECOM（欧洲通信研究院）；National Institute of Informatics（国立信息学研究所）；University of Eastern Finland（东芬兰大学）；Inria（法国国家信息与自动化研究所）；University of Edinburgh（爱丁堡大学）；NEC Corp.（日本电气股份有限公司）
CompSpoof: A Dataset and Joint Learning Framework for Component-Level Audio Anti-spoofing Countermeasures
Suzhou Municipal Key Laboratory of Multimodal Intelligent Systems（多模态智能系统苏州市重点实验室）；Digital Innovation Research Center（数字创新研究中心）；Duke Kunshan University（杜克昆山大学）；OfSpectrum, Inc.（OfSpectrum公司）
Genuine-Focused Learning using Mask AutoEncoder for Generalized Fake Audio Detection
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Tsinghua University（清华大学）；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）；Shanghai Polytechnic University（上海第二工业大学）
Boundary and Intra-Segment Learning for Partial Audio Deepfake Localization
Sun Yat-sen University（中山大学）；China Mobile Internet Corporation（中国移动互联网公司）；The Hong Kong Polytechnic University（香港理工大学）；Nanyang Technological University（南洋理工大学）
Wav2DF-TSL: Two-stage Learning with Efficient Pre-training and Hierarchical Experts Fusion for Robust Audio Deepfake Detection
School of Computer Science and Technology, Xinjiang University, Urumqi, China（新疆大学计算机科学与技术学院）；Hefei iFly Digital Technology Co. Ltd., Hefei, China（合肥iFly数字技术有限公司）；University of Science and Technology of China, Hefei, China（中国科学技术大学）；Department of Electronic Engineering, Tsinghua University, Beijing, China（清华大学电子工程系）
Fraunhofer AISEC（弗劳恩霍夫人工智能安全研究所）；Wrocław University of Science and Technology（沃斯克拉大学科学与技术学院）；Resemble AI（Resemble AI公司）；Technical University of Cluj-Napoca（克鲁克斯-纳波卡技术大学）；Neodyme AG（Neodyme公司）；TU Munich（慕尼黑技术大学）
Half-Truth: A Partially Fake Audio Detection Dataset
National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所模式识别国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）
本文构建 HAD 部分伪造音频检测数据集，采用最新语音合成技术仅替换语句中的少数词语，支持伪造检测与篡改区域定位，并通过基准实验证明其比完全伪造音频更具挑战性。
An Unsupervised Domain Adaptation Method for Locating Manipulated Region in partially fake Audio
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Department of Automation, Tsinghua University（清华大学自动化系）
Manipulated Regions Localization For Partially Deepfake Audio: A Survey
State Key Laboratory of Multi-modal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所，中国科学院）；Department of Automation, Tsinghua University（自动化系，清华大学）；Department of Automation and Beijing National Research Center for Information Science and Technology, Tsinghua University（自动化系和北京信息科学与技术国家研究中心，清华大学）；University of Chinese Academy of Sciences（中国科学院大学）
ADD 2022: the First Audio Deep Synthesis Detection Challenge
National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所模式识别国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；National University of Singapore（新加坡国立大学）
Temporal Variability and Multi-Viewed Self-Supervised Representations to Tackle the ASVspoof5 Deepfake Challenge
Communication University of China（中国传媒大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
ADD 2023: the Second Audio Deepfake Detection Challenge
Peoples' Friendship University of Russia (RUDN University)（俄罗斯人民友谊大学（RUDN大学））；Joint Institute for Nuclear Research（联合核研究所）；Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）；University of Skövde（舍夫德大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Tsinghua University（清华大学）；National University of Singapore（新加坡国立大学）；The Chinese University of Hong Kong（香港中文大学）
Deepfake Detection System for the ADD Challenge Track 3.2 Based on Score Fusion
Institute of Acoustics, Chinese Academy of Sciences（中国科学院声学研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Duke Kunshan University（昆山杜克大学）；Wuhan University（武汉大学）
Deep Spectro-temporal Artifacts for Detecting Synthesized Speech
Tianjin University（天津大学）；National Institute of Informatics（国立情报学研究所）；Taiyuan University of Technology（太原理工大学）；Japan Advanced Institute of Science and Technology（北陆先端科学技术大学院大学）
What You Train Is What You Get: Gender Bias, Training Composition, and Post-Hoc Mitigation in Audio Deepfake Detection
School of Computing, Wichita State University（威奇托州立大学计算学院）；Institut national de la recherche scientifique (INRS–EMT)（国家科学研究院（INRS–EMT））；INRS-UQO Mixed Research Unit on Cybersecurity（INRS-UQO网络安全混合研究单位）
MuSFA: Improving Music Structural Function Analysis with Partially Labeled Data
SAMI, ByteDance（字节跳动SAMI）
音乐结构分析（MSA）系统旨在将歌曲记录分割成具有有用标签的非重叠部分。以前的MSA系统通常在后处理步骤中预测抽象标签，并需要歌曲的完整上下文。相比之下，我们最近提出了一个被称为“音乐结构功能分析”（MuSFA）的监督框架，该框架直接从音频中建模和预测有意义的标签，如“诗句”和“合唱”，而不需要歌曲的完整上下文。然而，该系统的性能取决于训练数据的数量和质量。在本文中，我们建议重新利用公共数据集HookTheory Lead Sheet Dataset（HLSD）来提高性能。HLSD包含超过18K的音乐片段摘录，最初收集用于研究自动旋律协调。我们将每个摘录视为部分标记的歌曲并提供标签映射，以便HLSD可以与其他公共数据集（如SALAMI，RWC和Isophonics）一起使用。在交叉数据集评估中，我们发现在训练中包含HLSD可以将最先进的边界检测和切片标记分数分别提高约3%和约1%。
