{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AA7QQVI2SDMBZ2F7GPINWQ6TJS","short_pith_number":"pith:AA7QQVI2","schema_version":"1.0","canonical_sha256":"003f08551a90d81ce8bf33d0db43d34c9d9433cac4aa8ea75eac55c3618245aa","source":{"kind":"arxiv","id":"2411.17537","version":1},"attestation_state":"computed","paper":{"title":"Towards Maximum Likelihood Training for Transducer-based Streaming Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"eess.AS","authors_text":"Hyeonseung Lee, Ji Won Yoon, Nam Soo Kim, Sungsoo Kim","submitted_at":"2024-11-26T15:53:13Z","abstract_excerpt":"Transducer neural networks have emerged as the mainstream approach for streaming automatic speech recognition (ASR), offering state-of-the-art performance in balancing accuracy and latency. In the conventional framework, streaming transducer models are trained to maximize the likelihood function based on non-streaming recursion rules. However, this approach leads to a mismatch between training and inference, resulting in the issue of deformed likelihood and consequently suboptimal ASR accuracy. We introduce a mathematical quantification of the gap between the actual likelihood and the deformed"},"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":"2411.17537","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2024-11-26T15:53:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"29093bfbc7e1518f241ac8ca44eaefa33f7d5924f4d86da5114db41acde0bd46","abstract_canon_sha256":"d57f06ce26595bf5816b0ffb22b718e71898662c57d7177cf5f6cf1d92eacdff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:40:48.658043Z","signature_b64":"7qhNYJWtoB467ML+ghWZWAA6mq6VRNVIHQ7mnV+I08GyFZj8MSSu5MAvyEQ43AnDoKpKoJHLALk6EARgcbjwBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"003f08551a90d81ce8bf33d0db43d34c9d9433cac4aa8ea75eac55c3618245aa","last_reissued_at":"2026-07-05T09:40:48.657597Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:40:48.657597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Maximum Likelihood Training for Transducer-based Streaming Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"eess.AS","authors_text":"Hyeonseung Lee, Ji Won Yoon, Nam Soo Kim, Sungsoo Kim","submitted_at":"2024-11-26T15:53:13Z","abstract_excerpt":"Transducer neural networks have emerged as the mainstream approach for streaming automatic speech recognition (ASR), offering state-of-the-art performance in balancing accuracy and latency. In the conventional framework, streaming transducer models are trained to maximize the likelihood function based on non-streaming recursion rules. However, this approach leads to a mismatch between training and inference, resulting in the issue of deformed likelihood and consequently suboptimal ASR accuracy. We introduce a mathematical quantification of the gap between the actual likelihood and the deformed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17537","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/2411.17537/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":"2411.17537","created_at":"2026-07-05T09:40:48.657656+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.17537v1","created_at":"2026-07-05T09:40:48.657656+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17537","created_at":"2026-07-05T09:40:48.657656+00:00"},{"alias_kind":"pith_short_12","alias_value":"AA7QQVI2SDMB","created_at":"2026-07-05T09:40:48.657656+00:00"},{"alias_kind":"pith_short_16","alias_value":"AA7QQVI2SDMBZ2F7","created_at":"2026-07-05T09:40:48.657656+00:00"},{"alias_kind":"pith_short_8","alias_value":"AA7QQVI2","created_at":"2026-07-05T09:40:48.657656+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/AA7QQVI2SDMBZ2F7GPINWQ6TJS","json":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS.json","graph_json":"https://pith.science/api/pith-number/AA7QQVI2SDMBZ2F7GPINWQ6TJS/graph.json","events_json":"https://pith.science/api/pith-number/AA7QQVI2SDMBZ2F7GPINWQ6TJS/events.json","paper":"https://pith.science/paper/AA7QQVI2"},"agent_actions":{"view_html":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS","download_json":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS.json","view_paper":"https://pith.science/paper/AA7QQVI2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.17537&json=true","fetch_graph":"https://pith.science/api/pith-number/AA7QQVI2SDMBZ2F7GPINWQ6TJS/graph.json","fetch_events":"https://pith.science/api/pith-number/AA7QQVI2SDMBZ2F7GPINWQ6TJS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS/action/storage_attestation","attest_author":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS/action/author_attestation","sign_citation":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS/action/citation_signature","submit_replication":"https://pith.science/pith/AA7QQVI2SDMBZ2F7GPINWQ6TJS/action/replication_record"}},"created_at":"2026-07-05T09:40:48.657656+00:00","updated_at":"2026-07-05T09:40:48.657656+00:00"}