{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GVUHADJMUCNS7GKGCD26PB7W6E","short_pith_number":"pith:GVUHADJM","schema_version":"1.0","canonical_sha256":"3568700d2ca09b2f994610f5e787f6f12aa193b3c7eab3a4f277bc6f76861126","source":{"kind":"arxiv","id":"2404.07341","version":2},"attestation_state":"computed","paper":{"title":"Conformer-1: Robust ASR via Large-Scale Semisupervised Bootstrapping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Andrea Vanzo, Dillon Pulliam, Domenic Donato, Francis McCann Ramirez, Gabriel Oexle, Kevin Zhang, Luka Chkhetiani, Michael Liang, Michael Nguyen, Ruben Bousbib, Sergio Ramirez Martin, Taufiquzzaman Peyash, Yash Khare","submitted_at":"2024-04-10T20:40:24Z","abstract_excerpt":"This paper presents Conformer-1, an end-to-end Automatic Speech Recognition (ASR) model trained on an extensive dataset of 570k hours of speech audio data, 91% of which was acquired from publicly available sources. To achieve this, we perform Noisy Student Training after generating pseudo-labels for the unlabeled public data using a strong Conformer RNN-T baseline model. The addition of these pseudo-labeled data results in remarkable improvements in relative Word Error Rate (WER) by 11.5% and 24.3% for our asynchronous and realtime models, respectively. Additionally, the model is more robust t"},"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":"2404.07341","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2024-04-10T20:40:24Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SD"],"title_canon_sha256":"66caf5ffe8cfcf39f834fe51a7b2d94f9544ae24f291d314c40644553672a471","abstract_canon_sha256":"d983f1de9b784f520ef0a71f35960d70c373487028a34c821714031ed22644a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:37.314301Z","signature_b64":"CIUK3mW21HTyegfq2cusB/8O8bWb317Wwbqc2YF7ihebv9Cq2zq0OXFQcYrFkUgSPtbw67LbPjzeZr93mqULAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3568700d2ca09b2f994610f5e787f6f12aa193b3c7eab3a4f277bc6f76861126","last_reissued_at":"2026-07-05T08:07:37.313823Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:37.313823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conformer-1: Robust ASR via Large-Scale Semisupervised Bootstrapping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Andrea Vanzo, Dillon Pulliam, Domenic Donato, Francis McCann Ramirez, Gabriel Oexle, Kevin Zhang, Luka Chkhetiani, Michael Liang, Michael Nguyen, Ruben Bousbib, Sergio Ramirez Martin, Taufiquzzaman Peyash, Yash Khare","submitted_at":"2024-04-10T20:40:24Z","abstract_excerpt":"This paper presents Conformer-1, an end-to-end Automatic Speech Recognition (ASR) model trained on an extensive dataset of 570k hours of speech audio data, 91% of which was acquired from publicly available sources. To achieve this, we perform Noisy Student Training after generating pseudo-labels for the unlabeled public data using a strong Conformer RNN-T baseline model. The addition of these pseudo-labeled data results in remarkable improvements in relative Word Error Rate (WER) by 11.5% and 24.3% for our asynchronous and realtime models, respectively. Additionally, the model is more robust t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07341","kind":"arxiv","version":2},"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/2404.07341/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":"2404.07341","created_at":"2026-07-05T08:07:37.313884+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.07341v2","created_at":"2026-07-05T08:07:37.313884+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07341","created_at":"2026-07-05T08:07:37.313884+00:00"},{"alias_kind":"pith_short_12","alias_value":"GVUHADJMUCNS","created_at":"2026-07-05T08:07:37.313884+00:00"},{"alias_kind":"pith_short_16","alias_value":"GVUHADJMUCNS7GKG","created_at":"2026-07-05T08:07:37.313884+00:00"},{"alias_kind":"pith_short_8","alias_value":"GVUHADJM","created_at":"2026-07-05T08:07:37.313884+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02296","citing_title":"Spatial Speech Perception Systems: A Survey of Sound Source Localization, Directional Enhancement, and Speech Recognition","ref_index":85,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E","json":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E.json","graph_json":"https://pith.science/api/pith-number/GVUHADJMUCNS7GKGCD26PB7W6E/graph.json","events_json":"https://pith.science/api/pith-number/GVUHADJMUCNS7GKGCD26PB7W6E/events.json","paper":"https://pith.science/paper/GVUHADJM"},"agent_actions":{"view_html":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E","download_json":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E.json","view_paper":"https://pith.science/paper/GVUHADJM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.07341&json=true","fetch_graph":"https://pith.science/api/pith-number/GVUHADJMUCNS7GKGCD26PB7W6E/graph.json","fetch_events":"https://pith.science/api/pith-number/GVUHADJMUCNS7GKGCD26PB7W6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E/action/storage_attestation","attest_author":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E/action/author_attestation","sign_citation":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E/action/citation_signature","submit_replication":"https://pith.science/pith/GVUHADJMUCNS7GKGCD26PB7W6E/action/replication_record"}},"created_at":"2026-07-05T08:07:37.313884+00:00","updated_at":"2026-07-05T08:07:37.313884+00:00"}