{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:ZZ6URFP2CJ6OJXDPSUUFHGWGUQ","short_pith_number":"pith:ZZ6URFP2","schema_version":"1.0","canonical_sha256":"ce7d4895fa127ce4dc6f9528539ac6a41895beee958c9bc85e6ab0c6c8187b75","source":{"kind":"arxiv","id":"2012.10055","version":2},"attestation_state":"computed","paper":{"title":"End-to-End Speaker Diarization as Post-Processing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Kenji Nagamatsu, Paola Garcia, Shinji Watanabe, Shota Horiguchi, Yusuke Fujita","submitted_at":"2020-12-18T05:31:07Z","abstract_excerpt":"This paper investigates the utilization of an end-to-end diarization model as post-processing of conventional clustering-based diarization. Clustering-based diarization methods partition frames into clusters of the number of speakers; thus, they typically cannot handle overlapping speech because each frame is assigned to one speaker. On the other hand, some end-to-end diarization methods can handle overlapping speech by treating the problem as multi-label classification. Although some methods can treat a flexible number of speakers, they do not perform well when the number of speakers is large"},"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":"2012.10055","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-12-18T05:31:07Z","cross_cats_sorted":["cs.CL","cs.SD"],"title_canon_sha256":"8075b2945459713fb41e45dc1b4f364ec4470771c6f3d5fee37530efd5618f53","abstract_canon_sha256":"3d5c61cea74d743690f7f7b522590ff34f1be26acfaf62608a1e55c94659dbec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:01:33.569775Z","signature_b64":"H/zv4tNkh2lGEF53967KdwkC0RrT4i4iL7nfBXfMSJ0QQDfgsom+67vy67M2Y495oNsIVsBalqKMccNWuyIPAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce7d4895fa127ce4dc6f9528539ac6a41895beee958c9bc85e6ab0c6c8187b75","last_reissued_at":"2026-07-05T02:01:33.569268Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:01:33.569268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"End-to-End Speaker Diarization as Post-Processing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Kenji Nagamatsu, Paola Garcia, Shinji Watanabe, Shota Horiguchi, Yusuke Fujita","submitted_at":"2020-12-18T05:31:07Z","abstract_excerpt":"This paper investigates the utilization of an end-to-end diarization model as post-processing of conventional clustering-based diarization. Clustering-based diarization methods partition frames into clusters of the number of speakers; thus, they typically cannot handle overlapping speech because each frame is assigned to one speaker. On the other hand, some end-to-end diarization methods can handle overlapping speech by treating the problem as multi-label classification. Although some methods can treat a flexible number of speakers, they do not perform well when the number of speakers is large"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.10055","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/2012.10055/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":"2012.10055","created_at":"2026-07-05T02:01:33.569360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.10055v2","created_at":"2026-07-05T02:01:33.569360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.10055","created_at":"2026-07-05T02:01:33.569360+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZZ6URFP2CJ6O","created_at":"2026-07-05T02:01:33.569360+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZZ6URFP2CJ6OJXDP","created_at":"2026-07-05T02:01:33.569360+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZZ6URFP2","created_at":"2026-07-05T02:01:33.569360+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/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ","json":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ.json","graph_json":"https://pith.science/api/pith-number/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/graph.json","events_json":"https://pith.science/api/pith-number/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/events.json","paper":"https://pith.science/paper/ZZ6URFP2"},"agent_actions":{"view_html":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ","download_json":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ.json","view_paper":"https://pith.science/paper/ZZ6URFP2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.10055&json=true","fetch_graph":"https://pith.science/api/pith-number/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/action/storage_attestation","attest_author":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/action/author_attestation","sign_citation":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/action/citation_signature","submit_replication":"https://pith.science/pith/ZZ6URFP2CJ6OJXDPSUUFHGWGUQ/action/replication_record"}},"created_at":"2026-07-05T02:01:33.569360+00:00","updated_at":"2026-07-05T02:01:33.569360+00:00"}