{"paper_id":"0e8a7a15-618d-4e09-9d6d-a61ce051abc3","body_text":"Detection of Hidden Compression Artifacts in Re–Encoded Audio for Forensic Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Detection of Hidden Compression Artifacts in Re–Encoded Audio for Forensic Analysis Rahul Dixit, Deep Das, Anuja Dixit This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7336850/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract The growing use of audio on digital platforms and advances in editing tools have made verifying recording authenticity a pressing challenge in forensic and security contexts. A common obfuscation method re-encodes audio with lossy codecs (e.g., AAC, MP3) and converts it to a lossless format, removing metadata while leaving subtle compression artifacts. We address this by proposing a content-based authentication framework that bypasses file headers and analyzes intrinsic signal features. A balanced dataset derived from LibriSpeech is constructed with 5–second training and 2–second testing segments to simulate real–world re–encoding and assess temporal generalization. Our multi-perspective feature extraction combines wav2vec2 embeddings, spectral descriptors (MFCCs, mel spectrograms, chroma), and compression-artifact measures (phase coherence, temporal envelope dynamics). These feed into an optimized XGBoost–Random Forest ensemble with cross-validated voting weights. The system achieves 98.80% accuracy, perfect specificity, and zero false negatives for compressed audio, offering a scalable and reliable solution for covert manipulation detection where metadata-based methods fail. Audio forensics Compression artifact detection Ensemble learning Feature extraction wav2vec2 Digital audio authentication Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 12 Aug, 2025 Reviewers invited by journal 12 Aug, 2025 Editor assigned by journal 11 Aug, 2025 Submission checks completed at journal 11 Aug, 2025 First submitted to journal 10 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-7336850\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":499597958,\"identity\":\"a77e0c64-44b8-4643-8de1-3562481687c6\",\"order_by\":0,\"name\":\"Rahul Dixit\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIiWNgGAWjYFACHgaGB2wSDGwMzAcYEthAIowNYAk2fFoSQFrY2BJgWhobiNACUsFjAFMGtQYHMG8/e0wiocwisU++5+OHB2UM9ga3m9sfMNTYMfBJY9cpcyYvTSLhnIQxGxvvZiCDIXHDnYNAhx1LZmCTOYBViwRDjplEYpuEHFDLBiCDIUFyRiJQC9sBBjaJBOxa+N+AtfCwsfE8/gHUYg/R8g+PFgm4LTxsIFsY+yWAWhjb8Gl5Y2wB8UuaGYiRCNIyI7EvmQe3w3IMb3woq0uc33z48c0fZTb2bBLpDz58+GYnJz8DuxbM4ACDBHB8jYJRMApGwSggFwAA12ZPtMieKdAAAAAASUVORK5CYII=\",\"orcid\":\"\",\"institution\":\"Sardar Vallabhbhai National Institute of Technology\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Rahul\",\"middleName\":\"\",\"lastName\":\"Dixit\",\"suffix\":\"\"},{\"id\":499597959,\"identity\":\"110cc5f1-8ea5-4978-b489-3bfe031ebd2b\",\"order_by\":1,\"name\":\"Deep Das\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Sardar Vallabhbhai National Institute of Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Deep\",\"middleName\":\"\",\"lastName\":\"Das\",\"suffix\":\"\"},{\"id\":499597961,\"identity\":\"2112c99f-f0c5-4926-8fc2-1178b95283b0\",\"order_by\":2,\"name\":\"Anuja Dixit\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Motilal Nehru National Institute of Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Anuja\",\"middleName\":\"\",\"lastName\":\"Dixit\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-08-10 05:08:11\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-7336850/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-7336850/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":89047373,\"identity\":\"274c5bbb-bc84-461e-90f2-63d556880eee\",\"added_by\":\"auto\",\"created_at\":\"2025-08-14 07:00:47\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":980468,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"EncodingDetectioninAudioForgeryEdited10Aug.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7336850/v1_covered_a6ebb1ef-c4ec-4c80-9152-26e28abd6811.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Detection of Hidden Compression Artifacts in Re–Encoded Audio for Forensic Analysis\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":true,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":true,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"signal-image-and-video-processing\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"sivp\",\"sideBox\":\"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)\",\"snPcode\":\"11760\",\"submissionUrl\":\"https://submission.nature.com/new-submission/11760/3\",\"title\":\"Signal, Image and Video Processing\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"Springer Hybrid\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false},\"keywords\":\"Audio forensics, Compression artifact detection, Ensemble learning, Feature extraction, wav2vec2, Digital audio authentication\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-7336850/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-7336850/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"The growing use of audio on digital platforms and advances in editing tools have made verifying recording authenticity a pressing challenge in forensic and security contexts. A common obfuscation method re-encodes audio with lossy codecs (e.g., AAC, MP3) and converts it to a lossless format, removing metadata while leaving subtle compression artifacts. We address this by proposing a content-based authentication framework that bypasses file headers and analyzes intrinsic signal features. A balanced dataset derived from LibriSpeech is constructed with 5–second training and 2–second testing segments to simulate real–world re–encoding and assess temporal generalization. Our multi-perspective feature extraction combines wav2vec2 embeddings, spectral descriptors (MFCCs, mel spectrograms, chroma), and compression-artifact measures (phase coherence, temporal envelope dynamics). These feed into an optimized XGBoost–Random Forest ensemble with cross-validated voting weights. The system achieves 98.80% accuracy, perfect specificity, and zero false negatives for compressed audio, offering a scalable and reliable solution for covert manipulation detection where metadata-based methods fail.\",\"manuscriptTitle\":\"Detection of Hidden Compression Artifacts in Re–Encoded Audio for Forensic Analysis\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-08-14 06:52:40\",\"doi\":\"10.21203/rs.3.rs-7336850/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2025-08-12T21:10:45+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2025-08-12T21:10:03+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2025-08-11T11:46:09+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2025-08-11T11:45:13+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Signal, Image and Video Processing\",\"date\":\"2025-08-10T05:01:51+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"signal-image-and-video-processing\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"sivp\",\"sideBox\":\"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)\",\"snPcode\":\"11760\",\"submissionUrl\":\"https://submission.nature.com/new-submission/11760/3\",\"title\":\"Signal, Image and Video Processing\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"Springer Hybrid\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false}}],\"origin\":\"\",\"ownerIdentity\":\"37be4d4c-fb81-4efb-9830-62e8f7705f9f\",\"owner\":[],\"postedDate\":\"August 14th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2025-08-22T01:38:12+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-08-14 06:52:40\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-7336850\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-7336850\",\"identity\":\"rs-7336850\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}