{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NTVMJH2KOPEH3LJEVJZD5ZUHOJ","short_pith_number":"pith:NTVMJH2K","schema_version":"1.0","canonical_sha256":"6ceac49f4a73c87dad24aa723ee6877243936b9ba53a139e9dc8dd19cbbae5c3","source":{"kind":"arxiv","id":"2503.11026","version":2},"attestation_state":"computed","paper":{"title":"MAVFlow: Preserving Paralinguistic Elements with Conditional Flow Matching for Zero-Shot AV2AV Multilingual Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.MM"],"primary_cat":"eess.AS","authors_text":"Jeongsoo Choi, Se-Young Yun, Sungnyun Kim, Sungwoo Cho","submitted_at":"2025-03-14T02:48:43Z","abstract_excerpt":"Despite recent advances in text-to-speech (TTS) models, audio-visual-to-audio-visual (AV2AV) translation still faces a critical challenge: maintaining speaker consistency between the original and translated vocal and facial features. To address this issue, we propose a conditional flow matching (CFM) zero-shot audio-visual renderer that utilizes strong dual guidance from both audio and visual modalities. By leveraging multimodal guidance with CFM, our model robustly preserves speaker-specific characteristics and enhances zero-shot AV2AV translation abilities. For the audio modality, we enhance"},"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":"2503.11026","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-03-14T02:48:43Z","cross_cats_sorted":["cs.CV","cs.LG","cs.MM"],"title_canon_sha256":"0b0e5ce3bdaf7bf0905e56ecc9fe27e1cf457220192ebee61e53e7751ab5057b","abstract_canon_sha256":"44e9ed202a9d6cda40d55236e2eb4a929294d63ba46f3651c3686b195c24542f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:32.772205Z","signature_b64":"P/dVmdduauyNMXq1xyTbLHcfQb0sciUjpJif47pjBewMlb06HkKeatH5nI8viFrnfAJxDGPCeAEiHNvhlU1hBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ceac49f4a73c87dad24aa723ee6877243936b9ba53a139e9dc8dd19cbbae5c3","last_reissued_at":"2026-07-05T11:45:32.771574Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:32.771574Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAVFlow: Preserving Paralinguistic Elements with Conditional Flow Matching for Zero-Shot AV2AV Multilingual Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.MM"],"primary_cat":"eess.AS","authors_text":"Jeongsoo Choi, Se-Young Yun, Sungnyun Kim, Sungwoo Cho","submitted_at":"2025-03-14T02:48:43Z","abstract_excerpt":"Despite recent advances in text-to-speech (TTS) models, audio-visual-to-audio-visual (AV2AV) translation still faces a critical challenge: maintaining speaker consistency between the original and translated vocal and facial features. To address this issue, we propose a conditional flow matching (CFM) zero-shot audio-visual renderer that utilizes strong dual guidance from both audio and visual modalities. By leveraging multimodal guidance with CFM, our model robustly preserves speaker-specific characteristics and enhances zero-shot AV2AV translation abilities. For the audio modality, we enhance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.11026","kind":"arxiv","version":2},"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/2503.11026/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":"2503.11026","created_at":"2026-07-05T11:45:32.771658+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.11026v2","created_at":"2026-07-05T11:45:32.771658+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.11026","created_at":"2026-07-05T11:45:32.771658+00:00"},{"alias_kind":"pith_short_12","alias_value":"NTVMJH2KOPEH","created_at":"2026-07-05T11:45:32.771658+00:00"},{"alias_kind":"pith_short_16","alias_value":"NTVMJH2KOPEH3LJE","created_at":"2026-07-05T11:45:32.771658+00:00"},{"alias_kind":"pith_short_8","alias_value":"NTVMJH2K","created_at":"2026-07-05T11:45:32.771658+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02443","citing_title":"Breaking the Barriers of Text-Hungry and Audio-Deficient AI","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ","json":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ.json","graph_json":"https://pith.science/api/pith-number/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/graph.json","events_json":"https://pith.science/api/pith-number/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/events.json","paper":"https://pith.science/paper/NTVMJH2K"},"agent_actions":{"view_html":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ","download_json":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ.json","view_paper":"https://pith.science/paper/NTVMJH2K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.11026&json=true","fetch_graph":"https://pith.science/api/pith-number/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/graph.json","fetch_events":"https://pith.science/api/pith-number/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/action/storage_attestation","attest_author":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/action/author_attestation","sign_citation":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/action/citation_signature","submit_replication":"https://pith.science/pith/NTVMJH2KOPEH3LJEVJZD5ZUHOJ/action/replication_record"}},"created_at":"2026-07-05T11:45:32.771658+00:00","updated_at":"2026-07-05T11:45:32.771658+00:00"}