{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CZUCPDEO3M3EYWWLQMHO4ZV7H3","short_pith_number":"pith:CZUCPDEO","schema_version":"1.0","canonical_sha256":"1668278c8edb364c5acb830eee66bf3efdf6f9065c8cc2927a12cef9661d6f6d","source":{"kind":"arxiv","id":"2203.16776","version":4},"attestation_state":"computed","paper":{"title":"An Empirical Study of Language Model Integration for Transducer based Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"eess.AS","authors_text":"Chen Huang, Guanglu Wan, Huahuan Zheng, Ke Ding, Keyu An, Zhijian Ou","submitted_at":"2022-03-31T03:33:50Z","abstract_excerpt":"Utilizing text-only data with an external language model (ELM) in end-to-end RNN-Transducer (RNN-T) for speech recognition is challenging. Recently, a class of methods such as density ratio (DR) and internal language model estimation (ILME) have been developed, outperforming the classic shallow fusion (SF) method. The basic idea behind these methods is that RNN-T posterior should first subtract the implicitly learned internal language model (ILM) prior, in order to integrate the ELM. While recent studies suggest that RNN-T only learns some low-order language model information, the DR method us"},"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":"2203.16776","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2022-03-31T03:33:50Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"5aee1c4c9500c19749e06e1cc56037d5c5bd6499ccda8c895473b9833f8e81a5","abstract_canon_sha256":"2d8f529bc4811de13520b9e91f904414250ac1ef247b52f7cf1c65779f50e02b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:45:48.365733Z","signature_b64":"hEtdqJHa+blI8Ba7xulnF+7OwbIR2NBGSXAF/yTnWVPNlzgWUnuqaN57lRU1h1GNTWrye4enLoqr1Uqmk7CaBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1668278c8edb364c5acb830eee66bf3efdf6f9065c8cc2927a12cef9661d6f6d","last_reissued_at":"2026-07-05T04:45:48.365259Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:45:48.365259Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Empirical Study of Language Model Integration for Transducer based Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"eess.AS","authors_text":"Chen Huang, Guanglu Wan, Huahuan Zheng, Ke Ding, Keyu An, Zhijian Ou","submitted_at":"2022-03-31T03:33:50Z","abstract_excerpt":"Utilizing text-only data with an external language model (ELM) in end-to-end RNN-Transducer (RNN-T) for speech recognition is challenging. Recently, a class of methods such as density ratio (DR) and internal language model estimation (ILME) have been developed, outperforming the classic shallow fusion (SF) method. The basic idea behind these methods is that RNN-T posterior should first subtract the implicitly learned internal language model (ILM) prior, in order to integrate the ELM. While recent studies suggest that RNN-T only learns some low-order language model information, the DR method us"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.16776","kind":"arxiv","version":4},"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/2203.16776/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":"2203.16776","created_at":"2026-07-05T04:45:48.365324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.16776v4","created_at":"2026-07-05T04:45:48.365324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.16776","created_at":"2026-07-05T04:45:48.365324+00:00"},{"alias_kind":"pith_short_12","alias_value":"CZUCPDEO3M3E","created_at":"2026-07-05T04:45:48.365324+00:00"},{"alias_kind":"pith_short_16","alias_value":"CZUCPDEO3M3EYWWL","created_at":"2026-07-05T04:45:48.365324+00:00"},{"alias_kind":"pith_short_8","alias_value":"CZUCPDEO","created_at":"2026-07-05T04:45:48.365324+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/CZUCPDEO3M3EYWWLQMHO4ZV7H3","json":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3.json","graph_json":"https://pith.science/api/pith-number/CZUCPDEO3M3EYWWLQMHO4ZV7H3/graph.json","events_json":"https://pith.science/api/pith-number/CZUCPDEO3M3EYWWLQMHO4ZV7H3/events.json","paper":"https://pith.science/paper/CZUCPDEO"},"agent_actions":{"view_html":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3","download_json":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3.json","view_paper":"https://pith.science/paper/CZUCPDEO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.16776&json=true","fetch_graph":"https://pith.science/api/pith-number/CZUCPDEO3M3EYWWLQMHO4ZV7H3/graph.json","fetch_events":"https://pith.science/api/pith-number/CZUCPDEO3M3EYWWLQMHO4ZV7H3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3/action/storage_attestation","attest_author":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3/action/author_attestation","sign_citation":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3/action/citation_signature","submit_replication":"https://pith.science/pith/CZUCPDEO3M3EYWWLQMHO4ZV7H3/action/replication_record"}},"created_at":"2026-07-05T04:45:48.365324+00:00","updated_at":"2026-07-05T04:45:48.365324+00:00"}