{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JR34JEXHKN7RUJ2JTYX7VCB45P","short_pith_number":"pith:JR34JEXH","schema_version":"1.0","canonical_sha256":"4c77c492e7537f1a27499e2ffa883cebf6bfecf8ded84e55b088ad0494886405","source":{"kind":"arxiv","id":"2107.05233","version":1},"attestation_state":"computed","paper":{"title":"UniSpeech at scale: An Empirical Study of Pre-training Method on Large-Scale Speech Recognition Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Chengyi Wang, Furu Wei, Jinyu Li, Kenichi Kumatani, Shujie Liu, Yao Qian, Yu Wu","submitted_at":"2021-07-12T07:30:17Z","abstract_excerpt":"Recently, there has been a vast interest in self-supervised learning (SSL) where the model is pre-trained on large scale unlabeled data and then fine-tuned on a small labeled dataset. The common wisdom is that SSL helps resource-limited tasks in which only a limited amount of labeled data is available. The benefit of SSL keeps diminishing when the labeled training data amount increases. To our best knowledge, at most a few thousand hours of labeled data was used in the study of SSL. In contrast, the industry usually uses tens of thousands of hours of labeled data to build high-accuracy speech "},"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":"2107.05233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2021-07-12T07:30:17Z","cross_cats_sorted":[],"title_canon_sha256":"612d0385765f176249ae8465604aa6849035e023425a569e42faba312b9cb223","abstract_canon_sha256":"895852fb1b9a5ec29e325b80c2c115a414350f8db866bcb4eb4ff81d49567c02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:57:00.861226Z","signature_b64":"/4DiAMoCh5LkdCQnVBTjWcnpihglkX/Uc4nQRJILx81yOjty0vl9VDdHBY4geZhaKw/0sPwnmGDHNnvrGwI6Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c77c492e7537f1a27499e2ffa883cebf6bfecf8ded84e55b088ad0494886405","last_reissued_at":"2026-07-05T02:57:00.860793Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:57:00.860793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniSpeech at scale: An Empirical Study of Pre-training Method on Large-Scale Speech Recognition Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Chengyi Wang, Furu Wei, Jinyu Li, Kenichi Kumatani, Shujie Liu, Yao Qian, Yu Wu","submitted_at":"2021-07-12T07:30:17Z","abstract_excerpt":"Recently, there has been a vast interest in self-supervised learning (SSL) where the model is pre-trained on large scale unlabeled data and then fine-tuned on a small labeled dataset. The common wisdom is that SSL helps resource-limited tasks in which only a limited amount of labeled data is available. The benefit of SSL keeps diminishing when the labeled training data amount increases. To our best knowledge, at most a few thousand hours of labeled data was used in the study of SSL. In contrast, the industry usually uses tens of thousands of hours of labeled data to build high-accuracy speech "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05233","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/2107.05233/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":"2107.05233","created_at":"2026-07-05T02:57:00.860847+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.05233v1","created_at":"2026-07-05T02:57:00.860847+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05233","created_at":"2026-07-05T02:57:00.860847+00:00"},{"alias_kind":"pith_short_12","alias_value":"JR34JEXHKN7R","created_at":"2026-07-05T02:57:00.860847+00:00"},{"alias_kind":"pith_short_16","alias_value":"JR34JEXHKN7RUJ2J","created_at":"2026-07-05T02:57:00.860847+00:00"},{"alias_kind":"pith_short_8","alias_value":"JR34JEXH","created_at":"2026-07-05T02:57:00.860847+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/JR34JEXHKN7RUJ2JTYX7VCB45P","json":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P.json","graph_json":"https://pith.science/api/pith-number/JR34JEXHKN7RUJ2JTYX7VCB45P/graph.json","events_json":"https://pith.science/api/pith-number/JR34JEXHKN7RUJ2JTYX7VCB45P/events.json","paper":"https://pith.science/paper/JR34JEXH"},"agent_actions":{"view_html":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P","download_json":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P.json","view_paper":"https://pith.science/paper/JR34JEXH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.05233&json=true","fetch_graph":"https://pith.science/api/pith-number/JR34JEXHKN7RUJ2JTYX7VCB45P/graph.json","fetch_events":"https://pith.science/api/pith-number/JR34JEXHKN7RUJ2JTYX7VCB45P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P/action/storage_attestation","attest_author":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P/action/author_attestation","sign_citation":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P/action/citation_signature","submit_replication":"https://pith.science/pith/JR34JEXHKN7RUJ2JTYX7VCB45P/action/replication_record"}},"created_at":"2026-07-05T02:57:00.860847+00:00","updated_at":"2026-07-05T02:57:00.860847+00:00"}