{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XOYGMP2Q5H67P2DUXDL43VTIKO","short_pith_number":"pith:XOYGMP2Q","schema_version":"1.0","canonical_sha256":"bbb0663f50e9fdf7e874b8d7cdd668538567e34e1df2ae53764f5bdfca8332df","source":{"kind":"arxiv","id":"2109.08555","version":2},"attestation_state":"computed","paper":{"title":"Continuous Streaming Multi-Talker ASR with Dual-path Transducers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Desh Raj, Jinyu Li, Liang Lu, Yashesh Gaur, Zhuo Chen","submitted_at":"2021-09-17T13:57:52Z","abstract_excerpt":"Streaming recognition of multi-talker conversations has so far been evaluated only for 2-speaker single-turn sessions. In this paper, we investigate it for multi-turn meetings containing multiple speakers using the Streaming Unmixing and Recognition Transducer (SURT) model, and show that naively extending the single-turn model to this harder setting incurs a performance penalty. As a solution, we propose the dual-path (DP) modeling strategy first used for time-domain speech separation. We experiment with LSTM and Transformer based DP models, and show that they improve word error rate (WER) per"},"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":"2109.08555","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-17T13:57:52Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"8f64cdd2d127b0fad7dd062b2923549d8078b215c8fb31798cb6650a669151d0","abstract_canon_sha256":"2890b9f7470696df83f22cefb6ba94f4918021dd78dc31bcbee6ccb17fc38693"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:50:36.437450Z","signature_b64":"PQoH1iUBsmllQilYgrxD89pHIcoopfrvDUuu44spqK5pX9jDucdpUjmJXCS3BDwet3p3M4fmtTGvK8AY6NqKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bbb0663f50e9fdf7e874b8d7cdd668538567e34e1df2ae53764f5bdfca8332df","last_reissued_at":"2026-07-05T03:50:36.436894Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:50:36.436894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continuous Streaming Multi-Talker ASR with Dual-path Transducers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Desh Raj, Jinyu Li, Liang Lu, Yashesh Gaur, Zhuo Chen","submitted_at":"2021-09-17T13:57:52Z","abstract_excerpt":"Streaming recognition of multi-talker conversations has so far been evaluated only for 2-speaker single-turn sessions. In this paper, we investigate it for multi-turn meetings containing multiple speakers using the Streaming Unmixing and Recognition Transducer (SURT) model, and show that naively extending the single-turn model to this harder setting incurs a performance penalty. As a solution, we propose the dual-path (DP) modeling strategy first used for time-domain speech separation. We experiment with LSTM and Transformer based DP models, and show that they improve word error rate (WER) per"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08555","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/2109.08555/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":"2109.08555","created_at":"2026-07-05T03:50:36.436954+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.08555v2","created_at":"2026-07-05T03:50:36.436954+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08555","created_at":"2026-07-05T03:50:36.436954+00:00"},{"alias_kind":"pith_short_12","alias_value":"XOYGMP2Q5H67","created_at":"2026-07-05T03:50:36.436954+00:00"},{"alias_kind":"pith_short_16","alias_value":"XOYGMP2Q5H67P2DU","created_at":"2026-07-05T03:50:36.436954+00:00"},{"alias_kind":"pith_short_8","alias_value":"XOYGMP2Q","created_at":"2026-07-05T03:50:36.436954+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/XOYGMP2Q5H67P2DUXDL43VTIKO","json":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO.json","graph_json":"https://pith.science/api/pith-number/XOYGMP2Q5H67P2DUXDL43VTIKO/graph.json","events_json":"https://pith.science/api/pith-number/XOYGMP2Q5H67P2DUXDL43VTIKO/events.json","paper":"https://pith.science/paper/XOYGMP2Q"},"agent_actions":{"view_html":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO","download_json":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO.json","view_paper":"https://pith.science/paper/XOYGMP2Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.08555&json=true","fetch_graph":"https://pith.science/api/pith-number/XOYGMP2Q5H67P2DUXDL43VTIKO/graph.json","fetch_events":"https://pith.science/api/pith-number/XOYGMP2Q5H67P2DUXDL43VTIKO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO/action/storage_attestation","attest_author":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO/action/author_attestation","sign_citation":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO/action/citation_signature","submit_replication":"https://pith.science/pith/XOYGMP2Q5H67P2DUXDL43VTIKO/action/replication_record"}},"created_at":"2026-07-05T03:50:36.436954+00:00","updated_at":"2026-07-05T03:50:36.436954+00:00"}