{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SEIAYFPXSIOI4YRI63U6GCQVTR","short_pith_number":"pith:SEIAYFPX","schema_version":"1.0","canonical_sha256":"91100c15f7921c8e6228f6e9e30a159c5aa02040b284b47dd08b467ab8a2324b","source":{"kind":"arxiv","id":"2409.05601","version":1},"attestation_state":"computed","paper":{"title":"Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training for Enhanced Speech Recognition and Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Boris Ginsburg, Georg Kucsko, Hainan Xu, Jagadeesh Balam, Nithin Rao Koluguri, Oleksii Hrinchuk, Travis Bartley","submitted_at":"2024-09-09T13:35:52Z","abstract_excerpt":"This paper presents a new method for training sequence-to-sequence models for speech recognition and translation tasks. Instead of the traditional approach of training models on short segments containing only lowercase or partial punctuation and capitalization (PnC) sentences, we propose training on longer utterances that include complete sentences with proper punctuation and capitalization. We achieve this by using the FastConformer architecture which allows training 1 Billion parameter models with sequences up to 60 seconds long with full attention. However, while training with PnC enhances "},"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":"2409.05601","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2024-09-09T13:35:52Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5346b7a8da9d361174e68761a6db18258a2d255300481639fcbd1352cf51aea9","abstract_canon_sha256":"1b64e55c9fbfdc2c323822ab1c8b96764515abae231cf4846486187add56a4e0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:04:51.839518Z","signature_b64":"Ue/67SU66QYidrkO/oT17rja23xkj7Yt+LpK3gqJslHTF5nwHxShW+X11xN3recSrBoNmFnGnHMMV/Ssl7CZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91100c15f7921c8e6228f6e9e30a159c5aa02040b284b47dd08b467ab8a2324b","last_reissued_at":"2026-07-05T09:04:51.839030Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:04:51.839030Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training for Enhanced Speech Recognition and Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Boris Ginsburg, Georg Kucsko, Hainan Xu, Jagadeesh Balam, Nithin Rao Koluguri, Oleksii Hrinchuk, Travis Bartley","submitted_at":"2024-09-09T13:35:52Z","abstract_excerpt":"This paper presents a new method for training sequence-to-sequence models for speech recognition and translation tasks. Instead of the traditional approach of training models on short segments containing only lowercase or partial punctuation and capitalization (PnC) sentences, we propose training on longer utterances that include complete sentences with proper punctuation and capitalization. We achieve this by using the FastConformer architecture which allows training 1 Billion parameter models with sequences up to 60 seconds long with full attention. However, while training with PnC enhances "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.05601","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/2409.05601/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":"2409.05601","created_at":"2026-07-05T09:04:51.839087+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.05601v1","created_at":"2026-07-05T09:04:51.839087+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.05601","created_at":"2026-07-05T09:04:51.839087+00:00"},{"alias_kind":"pith_short_12","alias_value":"SEIAYFPXSIOI","created_at":"2026-07-05T09:04:51.839087+00:00"},{"alias_kind":"pith_short_16","alias_value":"SEIAYFPXSIOI4YRI","created_at":"2026-07-05T09:04:51.839087+00:00"},{"alias_kind":"pith_short_8","alias_value":"SEIAYFPX","created_at":"2026-07-05T09:04:51.839087+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/SEIAYFPXSIOI4YRI63U6GCQVTR","json":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR.json","graph_json":"https://pith.science/api/pith-number/SEIAYFPXSIOI4YRI63U6GCQVTR/graph.json","events_json":"https://pith.science/api/pith-number/SEIAYFPXSIOI4YRI63U6GCQVTR/events.json","paper":"https://pith.science/paper/SEIAYFPX"},"agent_actions":{"view_html":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR","download_json":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR.json","view_paper":"https://pith.science/paper/SEIAYFPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.05601&json=true","fetch_graph":"https://pith.science/api/pith-number/SEIAYFPXSIOI4YRI63U6GCQVTR/graph.json","fetch_events":"https://pith.science/api/pith-number/SEIAYFPXSIOI4YRI63U6GCQVTR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR/action/storage_attestation","attest_author":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR/action/author_attestation","sign_citation":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR/action/citation_signature","submit_replication":"https://pith.science/pith/SEIAYFPXSIOI4YRI63U6GCQVTR/action/replication_record"}},"created_at":"2026-07-05T09:04:51.839087+00:00","updated_at":"2026-07-05T09:04:51.839087+00:00"}