{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FPD4LHEXR5KYKNC2UJZCVBWNHT","short_pith_number":"pith:FPD4LHEX","schema_version":"1.0","canonical_sha256":"2bc7c59c978f5585345aa2722a86cd3cf8d51b1d10c6bfb4be8b35e3be2af09d","source":{"kind":"arxiv","id":"2606.22274","version":1},"attestation_state":"computed","paper":{"title":"From Speech to Text Corpora: Evaluating ASR-Based Data Acquisition for Low-Resource Fongbe and Hausa","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mahounan Pericles Adjovi, Prasenjit Mitra, Roald Eiselen, Victor Olufemi","submitted_at":"2026-06-20T23:51:55Z","abstract_excerpt":"Low-resource African languages lack text corpora needed for language model training. We investigate whether ASR pipelines can extend text resources for two typologically distinct West African languages: Fongbe (tonal, diacritic-rich) and Hausa (non-tonal). We fine-tune MMS-300M on a curated 12.3-hour Fongbe dataset, achieving 9.48% WER on the ALFFA benchmark - a 78% relative reduction from the prior 44.04% baseline - while preserving tonal diacritics critical to the language. For Hausa, we apply an existing fine-tuned Whisper-Small model. We catalog 1,553 YouTube videos (236 hours) and process"},"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":"2606.22274","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-20T23:51:55Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"077ead7b1ca701b110da20dd98aafb4756971804f4f1b91589b9380669163649","abstract_canon_sha256":"8df4439e296015792d5597143fedb6c04c600b16da3f27460e5c11aa770f0a65"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:33.646393Z","signature_b64":"EvUFvSl2nrA1xWtqTqLAEcuf9aUrwWlAKiiVmi0FVPZfS83ET9CgIp1ZgazeNkg+t4cYwHDv66Vo5kdo3OFABw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bc7c59c978f5585345aa2722a86cd3cf8d51b1d10c6bfb4be8b35e3be2af09d","last_reissued_at":"2026-06-23T02:13:33.645986Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:33.645986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Speech to Text Corpora: Evaluating ASR-Based Data Acquisition for Low-Resource Fongbe and Hausa","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mahounan Pericles Adjovi, Prasenjit Mitra, Roald Eiselen, Victor Olufemi","submitted_at":"2026-06-20T23:51:55Z","abstract_excerpt":"Low-resource African languages lack text corpora needed for language model training. We investigate whether ASR pipelines can extend text resources for two typologically distinct West African languages: Fongbe (tonal, diacritic-rich) and Hausa (non-tonal). We fine-tune MMS-300M on a curated 12.3-hour Fongbe dataset, achieving 9.48% WER on the ALFFA benchmark - a 78% relative reduction from the prior 44.04% baseline - while preserving tonal diacritics critical to the language. For Hausa, we apply an existing fine-tuned Whisper-Small model. We catalog 1,553 YouTube videos (236 hours) and process"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22274","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/2606.22274/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":"2606.22274","created_at":"2026-06-23T02:13:33.646045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22274v1","created_at":"2026-06-23T02:13:33.646045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22274","created_at":"2026-06-23T02:13:33.646045+00:00"},{"alias_kind":"pith_short_12","alias_value":"FPD4LHEXR5KY","created_at":"2026-06-23T02:13:33.646045+00:00"},{"alias_kind":"pith_short_16","alias_value":"FPD4LHEXR5KYKNC2","created_at":"2026-06-23T02:13:33.646045+00:00"},{"alias_kind":"pith_short_8","alias_value":"FPD4LHEX","created_at":"2026-06-23T02:13:33.646045+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/FPD4LHEXR5KYKNC2UJZCVBWNHT","json":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT.json","graph_json":"https://pith.science/api/pith-number/FPD4LHEXR5KYKNC2UJZCVBWNHT/graph.json","events_json":"https://pith.science/api/pith-number/FPD4LHEXR5KYKNC2UJZCVBWNHT/events.json","paper":"https://pith.science/paper/FPD4LHEX"},"agent_actions":{"view_html":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT","download_json":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT.json","view_paper":"https://pith.science/paper/FPD4LHEX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22274&json=true","fetch_graph":"https://pith.science/api/pith-number/FPD4LHEXR5KYKNC2UJZCVBWNHT/graph.json","fetch_events":"https://pith.science/api/pith-number/FPD4LHEXR5KYKNC2UJZCVBWNHT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT/action/storage_attestation","attest_author":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT/action/author_attestation","sign_citation":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT/action/citation_signature","submit_replication":"https://pith.science/pith/FPD4LHEXR5KYKNC2UJZCVBWNHT/action/replication_record"}},"created_at":"2026-06-23T02:13:33.646045+00:00","updated_at":"2026-06-23T02:13:33.646045+00:00"}