{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:KHZPCZN7IROTVVKIMO7BOUGZWH","short_pith_number":"pith:KHZPCZN7","schema_version":"1.0","canonical_sha256":"51f2f165bf445d3ad54863be1750d9b1de9e6b87d4d9b31399c321088aa2ebd6","source":{"kind":"arxiv","id":"2111.05948","version":3},"attestation_state":"computed","paper":{"title":"Scaling ASR Improves Zero and Few Shot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alex Xiao, Christian Fuegen, Duc Le, Frank Zhang, Gil Keren, Ozlem Kalinli, Weiyi Zheng, Yatharth Saraf","submitted_at":"2021-11-10T21:18:59Z","abstract_excerpt":"With 4.5 million hours of English speech from 10 different sources across 120 countries and models of up to 10 billion parameters, we explore the frontiers of scale for automatic speech recognition. We propose data selection techniques to efficiently scale training data to find the most valuable samples in massive datasets. To efficiently scale model sizes, we leverage various optimizations such as sparse transducer loss and model sharding. By training 1-10B parameter universal English ASR models, we push the limits of speech recognition performance across many domains. Furthermore, our models"},"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":"2111.05948","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-10T21:18:59Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"ed2b2d2f6246a74b918137e21ec2e0929c3259fef744218235e7961f45294fca","abstract_canon_sha256":"a4d0600e36ce2aeecb0817d52d6d83fef576c594dd0d74c634a570ca41bba530"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:35:34.184188Z","signature_b64":"L8ESFALLsE0U5r5zs/hwZpzde5l2l9n6mFtdg1m9YoVzXiE0BEPZFWfB3HjXrSGw+lCiUCAmuOU+tnYwMxVVAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51f2f165bf445d3ad54863be1750d9b1de9e6b87d4d9b31399c321088aa2ebd6","last_reissued_at":"2026-07-05T03:35:34.183832Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:35:34.183832Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling ASR Improves Zero and Few Shot Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alex Xiao, Christian Fuegen, Duc Le, Frank Zhang, Gil Keren, Ozlem Kalinli, Weiyi Zheng, Yatharth Saraf","submitted_at":"2021-11-10T21:18:59Z","abstract_excerpt":"With 4.5 million hours of English speech from 10 different sources across 120 countries and models of up to 10 billion parameters, we explore the frontiers of scale for automatic speech recognition. We propose data selection techniques to efficiently scale training data to find the most valuable samples in massive datasets. To efficiently scale model sizes, we leverage various optimizations such as sparse transducer loss and model sharding. By training 1-10B parameter universal English ASR models, we push the limits of speech recognition performance across many domains. Furthermore, our models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.05948","kind":"arxiv","version":3},"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/2111.05948/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":"2111.05948","created_at":"2026-07-05T03:35:34.183886+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.05948v3","created_at":"2026-07-05T03:35:34.183886+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.05948","created_at":"2026-07-05T03:35:34.183886+00:00"},{"alias_kind":"pith_short_12","alias_value":"KHZPCZN7IROT","created_at":"2026-07-05T03:35:34.183886+00:00"},{"alias_kind":"pith_short_16","alias_value":"KHZPCZN7IROTVVKI","created_at":"2026-07-05T03:35:34.183886+00:00"},{"alias_kind":"pith_short_8","alias_value":"KHZPCZN7","created_at":"2026-07-05T03:35:34.183886+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/KHZPCZN7IROTVVKIMO7BOUGZWH","json":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH.json","graph_json":"https://pith.science/api/pith-number/KHZPCZN7IROTVVKIMO7BOUGZWH/graph.json","events_json":"https://pith.science/api/pith-number/KHZPCZN7IROTVVKIMO7BOUGZWH/events.json","paper":"https://pith.science/paper/KHZPCZN7"},"agent_actions":{"view_html":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH","download_json":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH.json","view_paper":"https://pith.science/paper/KHZPCZN7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.05948&json=true","fetch_graph":"https://pith.science/api/pith-number/KHZPCZN7IROTVVKIMO7BOUGZWH/graph.json","fetch_events":"https://pith.science/api/pith-number/KHZPCZN7IROTVVKIMO7BOUGZWH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH/action/storage_attestation","attest_author":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH/action/author_attestation","sign_citation":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH/action/citation_signature","submit_replication":"https://pith.science/pith/KHZPCZN7IROTVVKIMO7BOUGZWH/action/replication_record"}},"created_at":"2026-07-05T03:35:34.183886+00:00","updated_at":"2026-07-05T03:35:34.183886+00:00"}