{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LVKEFJCC7P53QTJT2PV3GFJ2LU","short_pith_number":"pith:LVKEFJCC","schema_version":"1.0","canonical_sha256":"5d5442a442fbfbb84d33d3ebb3153a5d3c4ac3465aef258089b65eba8a88e41e","source":{"kind":"arxiv","id":"2606.08505","version":1},"attestation_state":"computed","paper":{"title":"Fast and Robust On-Device Speaker Diarization: Relative Minimum Cluster Size for Stride-Accelerated Pipelines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Fumiaki Yamaguchi","submitted_at":"2026-06-07T08:10:14Z","abstract_excerpt":"Speech applications such as meeting transcription and voice agents would benefit from on-device speaker diarization, but practical adoption is limited by inference cost. We study how far a Pyannote 3.1-based pipeline can be accelerated on consumer hardware (an RTX 5070 Ti GPU and an Apple M4 laptop) while preserving diarization error rate (DER). A simple recipe: coarser segmentation stride and per-chunk embedding, yields multi-fold speedups and is DER-neutral on AMI, but degrades sharply on in-the-wild data: on VoxConverse, DER rises from 0.075 to 0.113. We trace the failure to speaker under-c"},"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":"2606.08505","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-06-07T08:10:14Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"778dc8548389de425bd9a30e6fde29026afbd2fe13e3a9c77adff0996e66b2f7","abstract_canon_sha256":"df6719fe8b4d7e42e9fabb8a97af87361dfdb00ff9d3c8c43e4b5249a529edd4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:38.513790Z","signature_b64":"+INF+HaI+NgJh3vWGJ9LYMR/HNdBl9+qaTe0ZmhZw73HSMf48vAUUgo9DHEbNPj6fnHWq3z0lzyIDBsT0be2Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d5442a442fbfbb84d33d3ebb3153a5d3c4ac3465aef258089b65eba8a88e41e","last_reissued_at":"2026-06-09T01:05:38.513376Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:38.513376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast and Robust On-Device Speaker Diarization: Relative Minimum Cluster Size for Stride-Accelerated Pipelines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Fumiaki Yamaguchi","submitted_at":"2026-06-07T08:10:14Z","abstract_excerpt":"Speech applications such as meeting transcription and voice agents would benefit from on-device speaker diarization, but practical adoption is limited by inference cost. We study how far a Pyannote 3.1-based pipeline can be accelerated on consumer hardware (an RTX 5070 Ti GPU and an Apple M4 laptop) while preserving diarization error rate (DER). A simple recipe: coarser segmentation stride and per-chunk embedding, yields multi-fold speedups and is DER-neutral on AMI, but degrades sharply on in-the-wild data: on VoxConverse, DER rises from 0.075 to 0.113. We trace the failure to speaker under-c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08505","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/2606.08505/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":"2606.08505","created_at":"2026-06-09T01:05:38.513441+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08505v1","created_at":"2026-06-09T01:05:38.513441+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08505","created_at":"2026-06-09T01:05:38.513441+00:00"},{"alias_kind":"pith_short_12","alias_value":"LVKEFJCC7P53","created_at":"2026-06-09T01:05:38.513441+00:00"},{"alias_kind":"pith_short_16","alias_value":"LVKEFJCC7P53QTJT","created_at":"2026-06-09T01:05:38.513441+00:00"},{"alias_kind":"pith_short_8","alias_value":"LVKEFJCC","created_at":"2026-06-09T01:05:38.513441+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.26842","citing_title":"voxmap-studio: An open-source speaker diarization annotation tool with built-in cost instrumentation","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU","json":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU.json","graph_json":"https://pith.science/api/pith-number/LVKEFJCC7P53QTJT2PV3GFJ2LU/graph.json","events_json":"https://pith.science/api/pith-number/LVKEFJCC7P53QTJT2PV3GFJ2LU/events.json","paper":"https://pith.science/paper/LVKEFJCC"},"agent_actions":{"view_html":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU","download_json":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU.json","view_paper":"https://pith.science/paper/LVKEFJCC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08505&json=true","fetch_graph":"https://pith.science/api/pith-number/LVKEFJCC7P53QTJT2PV3GFJ2LU/graph.json","fetch_events":"https://pith.science/api/pith-number/LVKEFJCC7P53QTJT2PV3GFJ2LU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU/action/storage_attestation","attest_author":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU/action/author_attestation","sign_citation":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU/action/citation_signature","submit_replication":"https://pith.science/pith/LVKEFJCC7P53QTJT2PV3GFJ2LU/action/replication_record"}},"created_at":"2026-06-09T01:05:38.513441+00:00","updated_at":"2026-06-09T01:05:38.513441+00:00"}