{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WUP76UNY7GTTKOP3RSKMTH6DVK","short_pith_number":"pith:WUP76UNY","schema_version":"1.0","canonical_sha256":"b51fff51b8f9a73539fb8c94c99fc3aa8579135c7172d5289aec33b2d36aec68","source":{"kind":"arxiv","id":"2409.08628","version":1},"attestation_state":"computed","paper":{"title":"Rhythmic Foley: A Framework For Seamless Audio-Visual Alignment In Video-to-Audio Synthesis","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Dan Luo, Huan Liao, Jun Wang, Zhiheng Li, Zhiqi Huang, Zhiyong Wu","submitted_at":"2024-09-13T08:33:03Z","abstract_excerpt":"Our research introduces an innovative framework for video-to-audio synthesis, which solves the problems of audio-video desynchronization and semantic loss in the audio. By incorporating a semantic alignment adapter and a temporal synchronization adapter, our method significantly improves semantic integrity and the precision of beat point synchronization, particularly in fast-paced action sequences. Utilizing a contrastive audio-visual pre-trained encoder, our model is trained with video and high-quality audio data, improving the quality of the generated audio. This dual-adapter approach empowe"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2409.08628","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SD","submitted_at":"2024-09-13T08:33:03Z","cross_cats_sorted":["cs.MM","eess.AS"],"title_canon_sha256":"cd8af3ff0c91a5c26510dd593293e3d40947a648a396915dec33c2ec0daed437","abstract_canon_sha256":"566a91dc98bad7050d131a3c74b2d7ad6a483fed30e9f9e2e6c13082d37ceb60"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:06:44.658065Z","signature_b64":"JvKE+eLJAqjC7lz2UbFXBcghTvE2WDN+HVEK96JM7rLPOWQpH5n7EqoIn6beWXmZslvdGRjQhzzIVaaBMlTpCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b51fff51b8f9a73539fb8c94c99fc3aa8579135c7172d5289aec33b2d36aec68","last_reissued_at":"2026-07-05T09:06:44.657644Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:06:44.657644Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rhythmic Foley: A Framework For Seamless Audio-Visual Alignment In Video-to-Audio Synthesis","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Dan Luo, Huan Liao, Jun Wang, Zhiheng Li, Zhiqi Huang, Zhiyong Wu","submitted_at":"2024-09-13T08:33:03Z","abstract_excerpt":"Our research introduces an innovative framework for video-to-audio synthesis, which solves the problems of audio-video desynchronization and semantic loss in the audio. By incorporating a semantic alignment adapter and a temporal synchronization adapter, our method significantly improves semantic integrity and the precision of beat point synchronization, particularly in fast-paced action sequences. Utilizing a contrastive audio-visual pre-trained encoder, our model is trained with video and high-quality audio data, improving the quality of the generated audio. This dual-adapter approach empowe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.08628","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2409.08628/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2409.08628","created_at":"2026-07-05T09:06:44.657702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.08628v1","created_at":"2026-07-05T09:06:44.657702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.08628","created_at":"2026-07-05T09:06:44.657702+00:00"},{"alias_kind":"pith_short_12","alias_value":"WUP76UNY7GTT","created_at":"2026-07-05T09:06:44.657702+00:00"},{"alias_kind":"pith_short_16","alias_value":"WUP76UNY7GTTKOP3","created_at":"2026-07-05T09:06:44.657702+00:00"},{"alias_kind":"pith_short_8","alias_value":"WUP76UNY","created_at":"2026-07-05T09:06:44.657702+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK","json":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK.json","graph_json":"https://pith.science/api/pith-number/WUP76UNY7GTTKOP3RSKMTH6DVK/graph.json","events_json":"https://pith.science/api/pith-number/WUP76UNY7GTTKOP3RSKMTH6DVK/events.json","paper":"https://pith.science/paper/WUP76UNY"},"agent_actions":{"view_html":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK","download_json":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK.json","view_paper":"https://pith.science/paper/WUP76UNY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.08628&json=true","fetch_graph":"https://pith.science/api/pith-number/WUP76UNY7GTTKOP3RSKMTH6DVK/graph.json","fetch_events":"https://pith.science/api/pith-number/WUP76UNY7GTTKOP3RSKMTH6DVK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK/action/storage_attestation","attest_author":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK/action/author_attestation","sign_citation":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK/action/citation_signature","submit_replication":"https://pith.science/pith/WUP76UNY7GTTKOP3RSKMTH6DVK/action/replication_record"}},"created_at":"2026-07-05T09:06:44.657702+00:00","updated_at":"2026-07-05T09:06:44.657702+00:00"}