{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6IQJSMT7NXRJEOH644XYWNJF25","short_pith_number":"pith:6IQJSMT7","schema_version":"1.0","canonical_sha256":"f22099327f6de29238fee72f8b3525d74335dbde77580d82efc4c3edb0a003a6","source":{"kind":"arxiv","id":"2508.13618","version":1},"attestation_state":"computed","paper":{"title":"TalkVid: A Large-Scale Diversified Dataset for Audio-Driven Talking Head Synthesis","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Benyou Wang, Chenghao Zhu, Guanbin Li, Harry Yang, Hejin Huang, Junying Chen, Michael Guan, Pengcheng Chen, Rongsheng Wang, Ser-Nam Lim, Shunian Chen, Yexin Liu, Zihan Ye","submitted_at":"2025-08-19T08:31:15Z","abstract_excerpt":"Audio-driven talking head synthesis has achieved remarkable photorealism, yet state-of-the-art (SOTA) models exhibit a critical failure: they lack generalization to the full spectrum of human diversity in ethnicity, language, and age groups. We argue that this generalization gap is a direct symptom of limitations in existing training data, which lack the necessary scale, quality, and diversity. To address this challenge, we introduce TalkVid, a new large-scale, high-quality, and diverse dataset containing 1244 hours of video from 7729 unique speakers. TalkVid is curated through a principled, m"},"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":"2508.13618","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-19T08:31:15Z","cross_cats_sorted":[],"title_canon_sha256":"a696c5d48e1d620372d3480f1cb7bd6aa52f1bfb6318632c2c8f1fe9fbbaac44","abstract_canon_sha256":"356b537e2299105ff2da0cfb3bd91e3c85cfe5d1645fc6a8ce01423d10e14469"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:51.266481Z","signature_b64":"xOpBVQwH0mB1+9J9Vgu/Rt7+yPSYQgQ6rUMicAzsVFL7yyf9K8UcAS770s1IxKUet3jetTsCdd1FDrjQh6/6BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f22099327f6de29238fee72f8b3525d74335dbde77580d82efc4c3edb0a003a6","last_reissued_at":"2026-07-05T11:55:51.265932Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:51.265932Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TalkVid: A Large-Scale Diversified Dataset for Audio-Driven Talking Head Synthesis","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Benyou Wang, Chenghao Zhu, Guanbin Li, Harry Yang, Hejin Huang, Junying Chen, Michael Guan, Pengcheng Chen, Rongsheng Wang, Ser-Nam Lim, Shunian Chen, Yexin Liu, Zihan Ye","submitted_at":"2025-08-19T08:31:15Z","abstract_excerpt":"Audio-driven talking head synthesis has achieved remarkable photorealism, yet state-of-the-art (SOTA) models exhibit a critical failure: they lack generalization to the full spectrum of human diversity in ethnicity, language, and age groups. We argue that this generalization gap is a direct symptom of limitations in existing training data, which lack the necessary scale, quality, and diversity. To address this challenge, we introduce TalkVid, a new large-scale, high-quality, and diverse dataset containing 1244 hours of video from 7729 unique speakers. TalkVid is curated through a principled, m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.13618","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/2508.13618/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":"2508.13618","created_at":"2026-07-05T11:55:51.266014+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.13618v1","created_at":"2026-07-05T11:55:51.266014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.13618","created_at":"2026-07-05T11:55:51.266014+00:00"},{"alias_kind":"pith_short_12","alias_value":"6IQJSMT7NXRJ","created_at":"2026-07-05T11:55:51.266014+00:00"},{"alias_kind":"pith_short_16","alias_value":"6IQJSMT7NXRJEOH6","created_at":"2026-07-05T11:55:51.266014+00:00"},{"alias_kind":"pith_short_8","alias_value":"6IQJSMT7","created_at":"2026-07-05T11:55:51.266014+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11180","citing_title":"Lip Forcing: Few-Step Autoregressive Diffusion for Real-time Lip Synchronization","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03672","citing_title":"Foley-Omni: A Unified Multimodal Generation Model from Task-Level Audio Synthesis to Complete Video Soundtrack Generation","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24176","citing_title":"Loki: Representation over Architecture for Diffusion-Based Portrait Animation","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25659","citing_title":"StreamChar: Long-Horizon Streaming Character Audio-Video Generation with Decoupled Orchestration","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28056","citing_title":"CogPortrait: Fine-Grained Eye-Region Control in Portrait Animation via Hierarchical Agent Planning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2512.00336","citing_title":"MVAD: A Benchmark Dataset for Multimodal AI-Generated Video-Audio Detection","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27918","citing_title":"Generate Your Talking Avatar from Video Reference","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01638","citing_title":"Omni-Fake: Benchmarking Unified Multimodal Social Media Deepfake Detection","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02948","citing_title":"AsymTalker: Identity-Consistent Long-Term Talking Head Generation via Asymmetric Distillation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17211","citing_title":"EmbodiedHead: Real-Time Listening and Speaking Avatar for Conversational Agents","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25","json":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25.json","graph_json":"https://pith.science/api/pith-number/6IQJSMT7NXRJEOH644XYWNJF25/graph.json","events_json":"https://pith.science/api/pith-number/6IQJSMT7NXRJEOH644XYWNJF25/events.json","paper":"https://pith.science/paper/6IQJSMT7"},"agent_actions":{"view_html":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25","download_json":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25.json","view_paper":"https://pith.science/paper/6IQJSMT7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.13618&json=true","fetch_graph":"https://pith.science/api/pith-number/6IQJSMT7NXRJEOH644XYWNJF25/graph.json","fetch_events":"https://pith.science/api/pith-number/6IQJSMT7NXRJEOH644XYWNJF25/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25/action/storage_attestation","attest_author":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25/action/author_attestation","sign_citation":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25/action/citation_signature","submit_replication":"https://pith.science/pith/6IQJSMT7NXRJEOH644XYWNJF25/action/replication_record"}},"created_at":"2026-07-05T11:55:51.266014+00:00","updated_at":"2026-07-05T11:55:51.266014+00:00"}