{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HLAFSAUQFWYUQPJM7TGYIUCHMA","short_pith_number":"pith:HLAFSAUQ","schema_version":"1.0","canonical_sha256":"3ac05902902db1483d2cfccd8450476002871becbfde543b126d447fa795697a","source":{"kind":"arxiv","id":"2110.15836","version":2},"attestation_state":"computed","paper":{"title":"Combining Unsupervised and Text Augmented Semi-Supervised Learning for Low Resourced Autoregressive Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chak-Fai Li, Francis Keith, Matthew Snover, William Hartmann","submitted_at":"2021-10-29T14:59:18Z","abstract_excerpt":"Recent advances in unsupervised representation learning have demonstrated the impact of pretraining on large amounts of read speech. We adapt these techniques for domain adaptation in low-resource -- both in terms of data and compute -- conversational and broadcast domains. Moving beyond CTC, we pretrain state-of-the-art Conformer models in an unsupervised manner. While the unsupervised approach outperforms traditional semi-supervised training, the techniques are complementary. Combining the techniques is a 5% absolute improvement in WER, averaged over all conditions, compared to semi-supervis"},"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":"2110.15836","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-10-29T14:59:18Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"1e8dbffe90735afdb46af82925c2736b502104f77e0670a0a2248021c258ed6d","abstract_canon_sha256":"b4698e25482cb18af1e3655df168691870d4b824d7d87c5de286e793454628b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:56:04.140500Z","signature_b64":"J/qwYEsXTlADlU62u5XpJkdG5IACDzW1novEAeOQVlP5+qdJl/JzKEhAkwfK/8shZBC2aIx3qP9nB8BaP/wQDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ac05902902db1483d2cfccd8450476002871becbfde543b126d447fa795697a","last_reissued_at":"2026-07-05T03:56:04.140058Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:56:04.140058Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Combining Unsupervised and Text Augmented Semi-Supervised Learning for Low Resourced Autoregressive Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chak-Fai Li, Francis Keith, Matthew Snover, William Hartmann","submitted_at":"2021-10-29T14:59:18Z","abstract_excerpt":"Recent advances in unsupervised representation learning have demonstrated the impact of pretraining on large amounts of read speech. We adapt these techniques for domain adaptation in low-resource -- both in terms of data and compute -- conversational and broadcast domains. Moving beyond CTC, we pretrain state-of-the-art Conformer models in an unsupervised manner. While the unsupervised approach outperforms traditional semi-supervised training, the techniques are complementary. Combining the techniques is a 5% absolute improvement in WER, averaged over all conditions, compared to semi-supervis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.15836","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/2110.15836/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":"2110.15836","created_at":"2026-07-05T03:56:04.140113+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.15836v2","created_at":"2026-07-05T03:56:04.140113+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.15836","created_at":"2026-07-05T03:56:04.140113+00:00"},{"alias_kind":"pith_short_12","alias_value":"HLAFSAUQFWYU","created_at":"2026-07-05T03:56:04.140113+00:00"},{"alias_kind":"pith_short_16","alias_value":"HLAFSAUQFWYUQPJM","created_at":"2026-07-05T03:56:04.140113+00:00"},{"alias_kind":"pith_short_8","alias_value":"HLAFSAUQ","created_at":"2026-07-05T03:56:04.140113+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/HLAFSAUQFWYUQPJM7TGYIUCHMA","json":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA.json","graph_json":"https://pith.science/api/pith-number/HLAFSAUQFWYUQPJM7TGYIUCHMA/graph.json","events_json":"https://pith.science/api/pith-number/HLAFSAUQFWYUQPJM7TGYIUCHMA/events.json","paper":"https://pith.science/paper/HLAFSAUQ"},"agent_actions":{"view_html":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA","download_json":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA.json","view_paper":"https://pith.science/paper/HLAFSAUQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.15836&json=true","fetch_graph":"https://pith.science/api/pith-number/HLAFSAUQFWYUQPJM7TGYIUCHMA/graph.json","fetch_events":"https://pith.science/api/pith-number/HLAFSAUQFWYUQPJM7TGYIUCHMA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA/action/storage_attestation","attest_author":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA/action/author_attestation","sign_citation":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA/action/citation_signature","submit_replication":"https://pith.science/pith/HLAFSAUQFWYUQPJM7TGYIUCHMA/action/replication_record"}},"created_at":"2026-07-05T03:56:04.140113+00:00","updated_at":"2026-07-05T03:56:04.140113+00:00"}