{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:AY7CZ2DGKX7YIF7VR3B4MIDLUV","short_pith_number":"pith:AY7CZ2DG","schema_version":"1.0","canonical_sha256":"063e2ce86655ff8417f58ec3c6206ba558b482bc4fd24285b7664ea7d30be450","source":{"kind":"arxiv","id":"2306.14170","version":1},"attestation_state":"computed","paper":{"title":"AV-SepFormer: Cross-Attention SepFormer for Audio-Visual Target Speaker Extraction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Heinrich Dinkel, Helen Meng, Jiuxin Lin, Junbo Zhang, Jun Chen, Xinyu Cai, Yongqing Wang, Yujun Wang, Zhiyong Wu, Zhiyong Yan","submitted_at":"2023-06-25T08:31:12Z","abstract_excerpt":"Visual information can serve as an effective cue for target speaker extraction (TSE) and is vital to improving extraction performance. In this paper, we propose AV-SepFormer, a SepFormer-based attention dual-scale model that utilizes cross- and self-attention to fuse and model features from audio and visual. AV-SepFormer splits the audio feature into a number of chunks, equivalent to the length of the visual feature. Then self- and cross-attention are employed to model and fuse the multi-modal features. Furthermore, we use a novel 2D positional encoding, that introduces the positional informat"},"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":"2306.14170","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.MM","submitted_at":"2023-06-25T08:31:12Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"a270e81f3ce8d8dcc1335d5520ddbd61dfb5456fedcaa88eebf8cc1e84f9e529","abstract_canon_sha256":"ce48ac79f70de62b1abce4988d310a1cf1eb3452460b5c323af2510d1c61cd3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:24:32.458269Z","signature_b64":"a7mlRuqpwY9JypKhHfX7cmVAlJbZyZl5orRAA0TFqx/8VCFTMSwggd06/unIe6A3g8hlbczx3ypX12a+IgWeDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"063e2ce86655ff8417f58ec3c6206ba558b482bc4fd24285b7664ea7d30be450","last_reissued_at":"2026-07-05T06:24:32.457803Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:24:32.457803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AV-SepFormer: Cross-Attention SepFormer for Audio-Visual Target Speaker Extraction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Heinrich Dinkel, Helen Meng, Jiuxin Lin, Junbo Zhang, Jun Chen, Xinyu Cai, Yongqing Wang, Yujun Wang, Zhiyong Wu, Zhiyong Yan","submitted_at":"2023-06-25T08:31:12Z","abstract_excerpt":"Visual information can serve as an effective cue for target speaker extraction (TSE) and is vital to improving extraction performance. In this paper, we propose AV-SepFormer, a SepFormer-based attention dual-scale model that utilizes cross- and self-attention to fuse and model features from audio and visual. AV-SepFormer splits the audio feature into a number of chunks, equivalent to the length of the visual feature. Then self- and cross-attention are employed to model and fuse the multi-modal features. Furthermore, we use a novel 2D positional encoding, that introduces the positional informat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.14170","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/2306.14170/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":"2306.14170","created_at":"2026-07-05T06:24:32.457868+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.14170v1","created_at":"2026-07-05T06:24:32.457868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.14170","created_at":"2026-07-05T06:24:32.457868+00:00"},{"alias_kind":"pith_short_12","alias_value":"AY7CZ2DGKX7Y","created_at":"2026-07-05T06:24:32.457868+00:00"},{"alias_kind":"pith_short_16","alias_value":"AY7CZ2DGKX7YIF7V","created_at":"2026-07-05T06:24:32.457868+00:00"},{"alias_kind":"pith_short_8","alias_value":"AY7CZ2DG","created_at":"2026-07-05T06:24:32.457868+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/AY7CZ2DGKX7YIF7VR3B4MIDLUV","json":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV.json","graph_json":"https://pith.science/api/pith-number/AY7CZ2DGKX7YIF7VR3B4MIDLUV/graph.json","events_json":"https://pith.science/api/pith-number/AY7CZ2DGKX7YIF7VR3B4MIDLUV/events.json","paper":"https://pith.science/paper/AY7CZ2DG"},"agent_actions":{"view_html":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV","download_json":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV.json","view_paper":"https://pith.science/paper/AY7CZ2DG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.14170&json=true","fetch_graph":"https://pith.science/api/pith-number/AY7CZ2DGKX7YIF7VR3B4MIDLUV/graph.json","fetch_events":"https://pith.science/api/pith-number/AY7CZ2DGKX7YIF7VR3B4MIDLUV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV/action/storage_attestation","attest_author":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV/action/author_attestation","sign_citation":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV/action/citation_signature","submit_replication":"https://pith.science/pith/AY7CZ2DGKX7YIF7VR3B4MIDLUV/action/replication_record"}},"created_at":"2026-07-05T06:24:32.457868+00:00","updated_at":"2026-07-05T06:24:32.457868+00:00"}