{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MYFSYUHWZQPWS32ESFTBNJHYGS","short_pith_number":"pith:MYFSYUHW","schema_version":"1.0","canonical_sha256":"660b2c50f6cc1f696f44916616a4f834905cba8d369d35f999642710062ff900","source":{"kind":"arxiv","id":"2504.12254","version":2},"attestation_state":"computed","paper":{"title":"Advancing Arabic Speech Recognition Through Large-Scale Weakly Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hasan Abusheikh, Mahmoud Salhab, Marwan Elghitany, Mohammad Abusheikh, Shameed Sait, Syed Sibghat Ullah","submitted_at":"2025-04-16T17:05:14Z","abstract_excerpt":"Automatic speech recognition (ASR) is crucial for human-machine interaction in diverse applications like conversational agents, industrial robotics, call center automation, and automated subtitling. However, developing high-performance ASR models remains challenging, particularly for low-resource languages like Arabic, due to the scarcity of large, labeled speech datasets, which are costly and labor-intensive to produce. In this work, we employ weakly supervised learning to train an Arabic ASR model using the Conformer architecture. Our model is trained from scratch on 15,000 hours of weakly a"},"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":"2504.12254","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-16T17:05:14Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"90190e10e3eb8ad47927783a961c6e34c80d0335f356e10d002700dac21899cf","abstract_canon_sha256":"6884e56abbefc095aa76b3fcc84c3e613cdeab2f1ae6faca4c21ca4929465ff6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:37.621460Z","signature_b64":"eT4q2tEoRLBryLUDuPpYmFhmI0DhkIhaiflyLKSkxBFB0v1EKFFQldPQMm+i0EWmU7xDwcwh++I7FJS+hskiAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"660b2c50f6cc1f696f44916616a4f834905cba8d369d35f999642710062ff900","last_reissued_at":"2026-07-05T10:51:37.620964Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:37.620964Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advancing Arabic Speech Recognition Through Large-Scale Weakly Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hasan Abusheikh, Mahmoud Salhab, Marwan Elghitany, Mohammad Abusheikh, Shameed Sait, Syed Sibghat Ullah","submitted_at":"2025-04-16T17:05:14Z","abstract_excerpt":"Automatic speech recognition (ASR) is crucial for human-machine interaction in diverse applications like conversational agents, industrial robotics, call center automation, and automated subtitling. However, developing high-performance ASR models remains challenging, particularly for low-resource languages like Arabic, due to the scarcity of large, labeled speech datasets, which are costly and labor-intensive to produce. In this work, we employ weakly supervised learning to train an Arabic ASR model using the Conformer architecture. Our model is trained from scratch on 15,000 hours of weakly a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12254","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/2504.12254/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":"2504.12254","created_at":"2026-07-05T10:51:37.621023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12254v2","created_at":"2026-07-05T10:51:37.621023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12254","created_at":"2026-07-05T10:51:37.621023+00:00"},{"alias_kind":"pith_short_12","alias_value":"MYFSYUHWZQPW","created_at":"2026-07-05T10:51:37.621023+00:00"},{"alias_kind":"pith_short_16","alias_value":"MYFSYUHWZQPWS32E","created_at":"2026-07-05T10:51:37.621023+00:00"},{"alias_kind":"pith_short_8","alias_value":"MYFSYUHW","created_at":"2026-07-05T10:51:37.621023+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.02038","citing_title":"NADI 2025: The First Multidialectal Arabic Speech Processing Shared Task","ref_index":80,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS","json":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS.json","graph_json":"https://pith.science/api/pith-number/MYFSYUHWZQPWS32ESFTBNJHYGS/graph.json","events_json":"https://pith.science/api/pith-number/MYFSYUHWZQPWS32ESFTBNJHYGS/events.json","paper":"https://pith.science/paper/MYFSYUHW"},"agent_actions":{"view_html":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS","download_json":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS.json","view_paper":"https://pith.science/paper/MYFSYUHW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12254&json=true","fetch_graph":"https://pith.science/api/pith-number/MYFSYUHWZQPWS32ESFTBNJHYGS/graph.json","fetch_events":"https://pith.science/api/pith-number/MYFSYUHWZQPWS32ESFTBNJHYGS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS/action/storage_attestation","attest_author":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS/action/author_attestation","sign_citation":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS/action/citation_signature","submit_replication":"https://pith.science/pith/MYFSYUHWZQPWS32ESFTBNJHYGS/action/replication_record"}},"created_at":"2026-07-05T10:51:37.621023+00:00","updated_at":"2026-07-05T10:51:37.621023+00:00"}