{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VUT4VG2LCDVQMKDPMZXPWOSI54","short_pith_number":"pith:VUT4VG2L","schema_version":"1.0","canonical_sha256":"ad27ca9b4b10eb06286f666efb3a48ef09a3d54e404bc3411153cfc64506cb9d","source":{"kind":"arxiv","id":"2506.18055","version":1},"attestation_state":"computed","paper":{"title":"Face-Voice Association for Audiovisual Active Speaker Detection in Egocentric Recordings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Jason Clarke, Stefan Goetze, Yoshihiko Gotoh","submitted_at":"2025-06-22T14:43:37Z","abstract_excerpt":"Audiovisual active speaker detection (ASD) is conventionally performed by modelling the temporal synchronisation of acoustic and visual speech cues. In egocentric recordings, however, the efficacy of synchronisation-based methods is compromised by occlusions, motion blur, and adverse acoustic conditions. In this work, a novel framework is proposed that exclusively leverages cross-modal face-voice associations to determine speaker activity. An existing face-voice association model is integrated with a transformer-based encoder that aggregates facial identity information by dynamically weighting"},"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":"2506.18055","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2025-06-22T14:43:37Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"b5af44490d342a5c92a21174a35adc9ce8effd86f0a2659b407d15e20664656c","abstract_canon_sha256":"3c2bb1a814e9b6fe03ac16685c38d3c39140fba4a2434a6b8c5497d59c604747"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:27.903490Z","signature_b64":"QnvHdZFy75QY81sZ1UpQfa+M4Westy8Q8g61GwbvCmdtoJ3PH150xyv8x5CmXdq6nbVOANd1Nql/NuQ3JaFcAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad27ca9b4b10eb06286f666efb3a48ef09a3d54e404bc3411153cfc64506cb9d","last_reissued_at":"2026-07-05T11:25:27.902891Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:27.902891Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Face-Voice Association for Audiovisual Active Speaker Detection in Egocentric Recordings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Jason Clarke, Stefan Goetze, Yoshihiko Gotoh","submitted_at":"2025-06-22T14:43:37Z","abstract_excerpt":"Audiovisual active speaker detection (ASD) is conventionally performed by modelling the temporal synchronisation of acoustic and visual speech cues. In egocentric recordings, however, the efficacy of synchronisation-based methods is compromised by occlusions, motion blur, and adverse acoustic conditions. In this work, a novel framework is proposed that exclusively leverages cross-modal face-voice associations to determine speaker activity. An existing face-voice association model is integrated with a transformer-based encoder that aggregates facial identity information by dynamically weighting"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18055","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/2506.18055/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":"2506.18055","created_at":"2026-07-05T11:25:27.902961+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.18055v1","created_at":"2026-07-05T11:25:27.902961+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18055","created_at":"2026-07-05T11:25:27.902961+00:00"},{"alias_kind":"pith_short_12","alias_value":"VUT4VG2LCDVQ","created_at":"2026-07-05T11:25:27.902961+00:00"},{"alias_kind":"pith_short_16","alias_value":"VUT4VG2LCDVQMKDP","created_at":"2026-07-05T11:25:27.902961+00:00"},{"alias_kind":"pith_short_8","alias_value":"VUT4VG2L","created_at":"2026-07-05T11:25:27.902961+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/VUT4VG2LCDVQMKDPMZXPWOSI54","json":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54.json","graph_json":"https://pith.science/api/pith-number/VUT4VG2LCDVQMKDPMZXPWOSI54/graph.json","events_json":"https://pith.science/api/pith-number/VUT4VG2LCDVQMKDPMZXPWOSI54/events.json","paper":"https://pith.science/paper/VUT4VG2L"},"agent_actions":{"view_html":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54","download_json":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54.json","view_paper":"https://pith.science/paper/VUT4VG2L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.18055&json=true","fetch_graph":"https://pith.science/api/pith-number/VUT4VG2LCDVQMKDPMZXPWOSI54/graph.json","fetch_events":"https://pith.science/api/pith-number/VUT4VG2LCDVQMKDPMZXPWOSI54/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54/action/storage_attestation","attest_author":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54/action/author_attestation","sign_citation":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54/action/citation_signature","submit_replication":"https://pith.science/pith/VUT4VG2LCDVQMKDPMZXPWOSI54/action/replication_record"}},"created_at":"2026-07-05T11:25:27.902961+00:00","updated_at":"2026-07-05T11:25:27.902961+00:00"}