{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DXADP732UEL6Y7WPOYRLLL2XUL","short_pith_number":"pith:DXADP732","schema_version":"1.0","canonical_sha256":"1dc037ff7aa117ec7ecf7622b5af57a2fd6c2177cea62d9df444cd888e99f871","source":{"kind":"arxiv","id":"2308.11863","version":3},"attestation_state":"computed","paper":{"title":"KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methods","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Antoine Nzeyimana","submitted_at":"2023-08-23T01:44:28Z","abstract_excerpt":"Despite recent availability of large transcribed Kinyarwanda speech data, achieving robust speech recognition for Kinyarwanda is still challenging. In this work, we show that using self-supervised pre-training, following a simple curriculum schedule during fine-tuning and using semi-supervised learning to leverage large unlabelled speech data significantly improve speech recognition performance for Kinyarwanda. Our approach focuses on using public domain data only. A new studio-quality speech dataset is collected from a public website, then used to train a clean baseline model. The clean basel"},"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":"2308.11863","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"eess.AS","submitted_at":"2023-08-23T01:44:28Z","cross_cats_sorted":["cs.LG","cs.SD"],"title_canon_sha256":"70c23d1fe5f6d46249851be037e836e1a70b0da4f5f04e085a1bcfd98af00615","abstract_canon_sha256":"5a1fa6efe8dd531eee976d79a7c190c06e970c1728c0dd8eb938e5464839ffec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:10.031634Z","signature_b64":"OhgQ7kQVdeMnWUZPlRvJ3LJtE0VfwSkIGG7EpRg5W7Vb9edPXyym2lgvFGfixveMNsjihi4ObkW8wPJ+gAjUAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1dc037ff7aa117ec7ecf7622b5af57a2fd6c2177cea62d9df444cd888e99f871","last_reissued_at":"2026-07-05T07:51:10.031192Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:10.031192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methods","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Antoine Nzeyimana","submitted_at":"2023-08-23T01:44:28Z","abstract_excerpt":"Despite recent availability of large transcribed Kinyarwanda speech data, achieving robust speech recognition for Kinyarwanda is still challenging. In this work, we show that using self-supervised pre-training, following a simple curriculum schedule during fine-tuning and using semi-supervised learning to leverage large unlabelled speech data significantly improve speech recognition performance for Kinyarwanda. Our approach focuses on using public domain data only. A new studio-quality speech dataset is collected from a public website, then used to train a clean baseline model. The clean basel"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.11863","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/2308.11863/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":"2308.11863","created_at":"2026-07-05T07:51:10.031254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.11863v3","created_at":"2026-07-05T07:51:10.031254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.11863","created_at":"2026-07-05T07:51:10.031254+00:00"},{"alias_kind":"pith_short_12","alias_value":"DXADP732UEL6","created_at":"2026-07-05T07:51:10.031254+00:00"},{"alias_kind":"pith_short_16","alias_value":"DXADP732UEL6Y7WP","created_at":"2026-07-05T07:51:10.031254+00:00"},{"alias_kind":"pith_short_8","alias_value":"DXADP732","created_at":"2026-07-05T07:51:10.031254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31642","citing_title":"Tone-Conditioned Curriculum Learning for Low-Resource Bantu Speech Recognition","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL","json":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL.json","graph_json":"https://pith.science/api/pith-number/DXADP732UEL6Y7WPOYRLLL2XUL/graph.json","events_json":"https://pith.science/api/pith-number/DXADP732UEL6Y7WPOYRLLL2XUL/events.json","paper":"https://pith.science/paper/DXADP732"},"agent_actions":{"view_html":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL","download_json":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL.json","view_paper":"https://pith.science/paper/DXADP732","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.11863&json=true","fetch_graph":"https://pith.science/api/pith-number/DXADP732UEL6Y7WPOYRLLL2XUL/graph.json","fetch_events":"https://pith.science/api/pith-number/DXADP732UEL6Y7WPOYRLLL2XUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL/action/storage_attestation","attest_author":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL/action/author_attestation","sign_citation":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL/action/citation_signature","submit_replication":"https://pith.science/pith/DXADP732UEL6Y7WPOYRLLL2XUL/action/replication_record"}},"created_at":"2026-07-05T07:51:10.031254+00:00","updated_at":"2026-07-05T07:51:10.031254+00:00"}