{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TXJTRAA2LNTDXZYVNI5C6YVVCZ","short_pith_number":"pith:TXJTRAA2","schema_version":"1.0","canonical_sha256":"9dd338801a5b663be7156a3a2f62b516486b447afb08f7a1cc114920c874c4b7","source":{"kind":"arxiv","id":"2201.03313","version":1},"attestation_state":"computed","paper":{"title":"Cross-Modal ASR Post-Processing System for Error Correction and Utterance Rejection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chao Jin, Dian Gu, Hongwei Zhou, Jing Du, Qinbo Dong, Ru Wu, Shiliang Pu, Xin Qi","submitted_at":"2022-01-10T12:29:55Z","abstract_excerpt":"Although modern automatic speech recognition (ASR) systems can achieve high performance, they may produce errors that weaken readers' experience and do harm to downstream tasks. To improve the accuracy and reliability of ASR hypotheses, we propose a cross-modal post-processing system for speech recognizers, which 1) fuses acoustic features and textual features from different modalities, 2) joints a confidence estimator and an error corrector in multi-task learning fashion and 3) unifies error correction and utterance rejection modules. Compared with single-modal or single-task models, our prop"},"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":"2201.03313","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-10T12:29:55Z","cross_cats_sorted":["cs.AI","cs.SD"],"title_canon_sha256":"60a37f045660b8804c961d1f3a0c9ace2a9ab2fce600dfec0131e40efa94f762","abstract_canon_sha256":"f6477e8ca72218635425941b20ece759dc6d97695e011449f0a70ed38bec1a73"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:47:08.422801Z","signature_b64":"+BuDsvadB1+wmXHEe8gGGmRmfqZEddAhEowI+UdPlhvx7eRdKCesv2bXH6RA5pmqZ+buh3e5jOuOZOmaM3IhAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dd338801a5b663be7156a3a2f62b516486b447afb08f7a1cc114920c874c4b7","last_reissued_at":"2026-07-05T03:47:08.422300Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:47:08.422300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cross-Modal ASR Post-Processing System for Error Correction and Utterance Rejection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chao Jin, Dian Gu, Hongwei Zhou, Jing Du, Qinbo Dong, Ru Wu, Shiliang Pu, Xin Qi","submitted_at":"2022-01-10T12:29:55Z","abstract_excerpt":"Although modern automatic speech recognition (ASR) systems can achieve high performance, they may produce errors that weaken readers' experience and do harm to downstream tasks. To improve the accuracy and reliability of ASR hypotheses, we propose a cross-modal post-processing system for speech recognizers, which 1) fuses acoustic features and textual features from different modalities, 2) joints a confidence estimator and an error corrector in multi-task learning fashion and 3) unifies error correction and utterance rejection modules. Compared with single-modal or single-task models, our prop"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.03313","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/2201.03313/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":"2201.03313","created_at":"2026-07-05T03:47:08.422373+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.03313v1","created_at":"2026-07-05T03:47:08.422373+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.03313","created_at":"2026-07-05T03:47:08.422373+00:00"},{"alias_kind":"pith_short_12","alias_value":"TXJTRAA2LNTD","created_at":"2026-07-05T03:47:08.422373+00:00"},{"alias_kind":"pith_short_16","alias_value":"TXJTRAA2LNTDXZYV","created_at":"2026-07-05T03:47:08.422373+00:00"},{"alias_kind":"pith_short_8","alias_value":"TXJTRAA2","created_at":"2026-07-05T03:47:08.422373+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.07285","citing_title":"Non-Intrusive Automatic Speech Recognition Refinement: A Survey","ref_index":179,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ","json":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ.json","graph_json":"https://pith.science/api/pith-number/TXJTRAA2LNTDXZYVNI5C6YVVCZ/graph.json","events_json":"https://pith.science/api/pith-number/TXJTRAA2LNTDXZYVNI5C6YVVCZ/events.json","paper":"https://pith.science/paper/TXJTRAA2"},"agent_actions":{"view_html":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ","download_json":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ.json","view_paper":"https://pith.science/paper/TXJTRAA2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.03313&json=true","fetch_graph":"https://pith.science/api/pith-number/TXJTRAA2LNTDXZYVNI5C6YVVCZ/graph.json","fetch_events":"https://pith.science/api/pith-number/TXJTRAA2LNTDXZYVNI5C6YVVCZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ/action/storage_attestation","attest_author":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ/action/author_attestation","sign_citation":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ/action/citation_signature","submit_replication":"https://pith.science/pith/TXJTRAA2LNTDXZYVNI5C6YVVCZ/action/replication_record"}},"created_at":"2026-07-05T03:47:08.422373+00:00","updated_at":"2026-07-05T03:47:08.422373+00:00"}