{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/digital-audio-tampering-detection-based-on-1","title":"Digital audio tampering detection based on spatio-temporal representation learning of electrical network frequency.","arxiv_id":null,"date":"2024-03-27","proceeding":"Springer 2024 3","authors":["Chunyan Zeng","Shuai Kong","Zhifeng Wang","Xiangkui Wan","Yunfan Chen","Kun Li","Yuhao Zhao"],"abstract":"The majority of Digital Audio Tampering Detection (DATD) methods, which are based on\r\nElectrical Network Frequency (ENF), predominantly concentrate on the static spatial infor-\r\nmation of ENF. Unfortunately, this focus neglects the temporal variation present in the ENF\r\ntime series. This limitation significantly hampers the ENF feature representation capability,\r\nconsequently diminishing the overall accuracy of tampering detection. To address this gap,\r\nour paper introduces an innovative digital audio tampering detection method founded on ENF\r\nspatio-temporal feature representation learning. To enhance the feature representation capa-\r\nbility and subsequently improve tampering detection accuracy, we propose the construction\r\nof a parallel spatio-temporal network model. This model incorporates both Convolutional\r\nNeural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) network\r\narchitectures. Through this hybrid model, we aim to deeply extract both ENF spatial and\r\ntemporal feature information. In the process of extracting spatial and temporal features of\r\nENF, we utilize high-precision Discrete Fourier Transform (DFT) analysis on digital audio.\r\nThis analysis allows us to extract ENF phase sequences, which are then adaptively divided\r\ninto frames through frame shifting. The result is feature matrices of uniform size, effectively\r\nrepresenting the spatial features of ENF. Concurrently, phase sequences are segmented into\r\nframes based on ENF time changes to capture the temporal features of ENF. Subsequently,\r\ndeep spatial and temporal features are extracted using CNN and BiLSTM, respectively. To\r\nfurther enhance the representation capability of the spatio-temporal features, we introduce\r\nan attention mechanism. This mechanism dynamically assigns weights to the deep spatial\r\nand temporal features, providing a nuanced and refined representation. Finally, a deep neural\r\nnetwork is employed to discern whether the audio has undergone tampering. Our experi-\r\nmental results validate the effectiveness of our approach, showcasing superior performance\r\ncompared to six state-of-the-art methods across three public databases for digital audio tam-\r\npering detection. This comprehensive methodology, focusing on both spatial and temporal\r\naspects of ENF, establishes a robust foundation for advancing the field of DATD and con-\r\ntributes significantly to improving detection accuracy.","url_abs":"https://doi.org/10.1007/s11042-024-18887-5","url_pdf":"https://doi.org/10.1007/s11042-024-18887-5","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"digital-audio-tampering-detection-based-on-1","repo_url":"https://github.com/JAYAKRISHNAN2712/Digital-audio-tampering-detection-based-on-spatio-temporal-representation-learning-of-ENF","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}