{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:OTM4K6P3J64LGJ7OXU5UP45CRC","short_pith_number":"pith:OTM4K6P3","schema_version":"1.0","canonical_sha256":"74d9c579fb4fb8b327eebd3b47f3a28898a8cf961defad533a12ad93d4303437","source":{"kind":"arxiv","id":"2110.15222","version":1},"attestation_state":"computed","paper":{"title":"Word-level confidence estimation for RNN transducers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Hagen Soltau, Izhak Shafran, Laurent El Shafey, Mingqiu Wang","submitted_at":"2021-09-28T18:38:00Z","abstract_excerpt":"Confidence estimate is an often requested feature in applications such as medical transcription where errors can impact patient care and the confidence estimate could be used to alert medical professionals to verify potential errors in recognition.\n  In this paper, we present a lightweight neural confidence model tailored for Automatic Speech Recognition (ASR) system with Recurrent Neural Network Transducers (RNN-T). Compared to other existing approaches, our model utilizes: (a) the time information associated with recognized words, which reduces the computational complexity, and (b) a simple "},"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":"2110.15222","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-28T18:38:00Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"9b5953061a65e5fed1ff37bdcb1299b89b4b3062feabe342bbfbe84233eec88c","abstract_canon_sha256":"a443908c91b1ee97087ab9a42de06d1d6d3c437fce96f15c0aed605a21073588"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:26:53.102096Z","signature_b64":"N0bbmiAO6Bp0HlgUjQV8vtBUhaM9ZvsNTuYW89GlSMXoytpe3JfbCDgFz7m/JZp1pS3HgoIk9krSl+DvtgGKCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74d9c579fb4fb8b327eebd3b47f3a28898a8cf961defad533a12ad93d4303437","last_reissued_at":"2026-07-05T03:26:53.100716Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:26:53.100716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Word-level confidence estimation for RNN transducers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Hagen Soltau, Izhak Shafran, Laurent El Shafey, Mingqiu Wang","submitted_at":"2021-09-28T18:38:00Z","abstract_excerpt":"Confidence estimate is an often requested feature in applications such as medical transcription where errors can impact patient care and the confidence estimate could be used to alert medical professionals to verify potential errors in recognition.\n  In this paper, we present a lightweight neural confidence model tailored for Automatic Speech Recognition (ASR) system with Recurrent Neural Network Transducers (RNN-T). Compared to other existing approaches, our model utilizes: (a) the time information associated with recognized words, which reduces the computational complexity, and (b) a simple "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.15222","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/2110.15222/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":"2110.15222","created_at":"2026-07-05T03:26:53.101276+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.15222v1","created_at":"2026-07-05T03:26:53.101276+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.15222","created_at":"2026-07-05T03:26:53.101276+00:00"},{"alias_kind":"pith_short_12","alias_value":"OTM4K6P3J64L","created_at":"2026-07-05T03:26:53.101276+00:00"},{"alias_kind":"pith_short_16","alias_value":"OTM4K6P3J64LGJ7O","created_at":"2026-07-05T03:26:53.101276+00:00"},{"alias_kind":"pith_short_8","alias_value":"OTM4K6P3","created_at":"2026-07-05T03:26:53.101276+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25404","citing_title":"Proactive for Uncertainty: Cause-Aware Error Diagnosis and Interactive Clarification for Spoken Dialogue Systems","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC","json":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC.json","graph_json":"https://pith.science/api/pith-number/OTM4K6P3J64LGJ7OXU5UP45CRC/graph.json","events_json":"https://pith.science/api/pith-number/OTM4K6P3J64LGJ7OXU5UP45CRC/events.json","paper":"https://pith.science/paper/OTM4K6P3"},"agent_actions":{"view_html":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC","download_json":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC.json","view_paper":"https://pith.science/paper/OTM4K6P3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.15222&json=true","fetch_graph":"https://pith.science/api/pith-number/OTM4K6P3J64LGJ7OXU5UP45CRC/graph.json","fetch_events":"https://pith.science/api/pith-number/OTM4K6P3J64LGJ7OXU5UP45CRC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC/action/storage_attestation","attest_author":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC/action/author_attestation","sign_citation":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC/action/citation_signature","submit_replication":"https://pith.science/pith/OTM4K6P3J64LGJ7OXU5UP45CRC/action/replication_record"}},"created_at":"2026-07-05T03:26:53.101276+00:00","updated_at":"2026-07-05T03:26:53.101276+00:00"}