{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:AHMMMXLHMX42CL2DINZNL3LGZE","short_pith_number":"pith:AHMMMXLH","schema_version":"1.0","canonical_sha256":"01d8c65d6765f9a12f434372d5ed66c9020b83c21c043511fe2f71ddf1e85dc8","source":{"kind":"arxiv","id":"2305.05084","version":6},"attestation_state":"computed","paper":{"title":"Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Ankur Kumar, Boris Ginsburg, Dima Rekesh, He Huang, Jagadeesh Balam, Krishna Puvvada, Nithin Rao Koluguri, Oleksii Hrinchuk, Samuel Kriman, Somshubra Majumdar, Vahid Noroozi","submitted_at":"2023-05-08T22:54:07Z","abstract_excerpt":"Conformer-based models have become the dominant end-to-end architecture for speech processing tasks. With the objective of enhancing the conformer architecture for efficient training and inference, we carefully redesigned Conformer with a novel downsampling schema. The proposed model, named Fast Conformer(FC), is 2.8x faster than the original Conformer, supports scaling to Billion parameters without any changes to the core architecture and also achieves state-of-the-art accuracy on Automatic Speech Recognition benchmarks. To enable transcription of long-form speech up to 11 hours, we replaced "},"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":"2305.05084","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2023-05-08T22:54:07Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"a3d5eb5dc2be628db888015859baef8a4108849469c9071c698b4e7e11ac19d1","abstract_canon_sha256":"c370117bc4980d6baa3d79eae9ca1ba1b29c5d0a73e0e495920b1da651f6a39e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:55:54.652848Z","signature_b64":"usiMSB3OcyGX53ru4N7SsPWJNhUK2d18P293peqMGGrV/JJ6+UKjYV6Vu7GHEWT2jgaife+3yqhnYm7l4npADg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01d8c65d6765f9a12f434372d5ed66c9020b83c21c043511fe2f71ddf1e85dc8","last_reissued_at":"2026-07-05T06:55:54.652348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:55:54.652348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Ankur Kumar, Boris Ginsburg, Dima Rekesh, He Huang, Jagadeesh Balam, Krishna Puvvada, Nithin Rao Koluguri, Oleksii Hrinchuk, Samuel Kriman, Somshubra Majumdar, Vahid Noroozi","submitted_at":"2023-05-08T22:54:07Z","abstract_excerpt":"Conformer-based models have become the dominant end-to-end architecture for speech processing tasks. With the objective of enhancing the conformer architecture for efficient training and inference, we carefully redesigned Conformer with a novel downsampling schema. The proposed model, named Fast Conformer(FC), is 2.8x faster than the original Conformer, supports scaling to Billion parameters without any changes to the core architecture and also achieves state-of-the-art accuracy on Automatic Speech Recognition benchmarks. To enable transcription of long-form speech up to 11 hours, we replaced "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.05084","kind":"arxiv","version":6},"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/2305.05084/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":"2305.05084","created_at":"2026-07-05T06:55:54.652419+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.05084v6","created_at":"2026-07-05T06:55:54.652419+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.05084","created_at":"2026-07-05T06:55:54.652419+00:00"},{"alias_kind":"pith_short_12","alias_value":"AHMMMXLHMX42","created_at":"2026-07-05T06:55:54.652419+00:00"},{"alias_kind":"pith_short_16","alias_value":"AHMMMXLHMX42CL2D","created_at":"2026-07-05T06:55:54.652419+00:00"},{"alias_kind":"pith_short_8","alias_value":"AHMMMXLH","created_at":"2026-07-05T06:55:54.652419+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08486","citing_title":"TRADE: Transducer-Augmented Decoder for Speech LLM","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31469","citing_title":"Scaling Conversational Hungarian ASR: The BEA-Dialogue+ Corpus","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE","json":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE.json","graph_json":"https://pith.science/api/pith-number/AHMMMXLHMX42CL2DINZNL3LGZE/graph.json","events_json":"https://pith.science/api/pith-number/AHMMMXLHMX42CL2DINZNL3LGZE/events.json","paper":"https://pith.science/paper/AHMMMXLH"},"agent_actions":{"view_html":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE","download_json":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE.json","view_paper":"https://pith.science/paper/AHMMMXLH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.05084&json=true","fetch_graph":"https://pith.science/api/pith-number/AHMMMXLHMX42CL2DINZNL3LGZE/graph.json","fetch_events":"https://pith.science/api/pith-number/AHMMMXLHMX42CL2DINZNL3LGZE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE/action/storage_attestation","attest_author":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE/action/author_attestation","sign_citation":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE/action/citation_signature","submit_replication":"https://pith.science/pith/AHMMMXLHMX42CL2DINZNL3LGZE/action/replication_record"}},"created_at":"2026-07-05T06:55:54.652419+00:00","updated_at":"2026-07-05T06:55:54.652419+00:00"}