{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ANFJDLW7JUORCBP3BVAKG2QGBJ","short_pith_number":"pith:ANFJDLW7","schema_version":"1.0","canonical_sha256":"034a91aedf4d1d1105fb0d40a36a060a452904869bb8e76884f6bde141d5ca3e","source":{"kind":"arxiv","id":"2406.06371","version":5},"attestation_state":"computed","paper":{"title":"mHuBERT-147: A Compact Multilingual HuBERT Model","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ioan Calapodescu, Laurent Besacier, Marcely Zanon Boito, Nikolaos Lagos, Vivek Iyer","submitted_at":"2024-06-10T15:32:42Z","abstract_excerpt":"We present mHuBERT-147, the first general-purpose massively multilingual HuBERT speech representation model trained on 90K hours of clean, open-license data. To scale up the multi-iteration HuBERT approach, we use faiss-based clustering, achieving 5.2x faster label assignment than the original method. We also apply a new multilingual batching up-sampling strategy, leveraging both language and dataset diversity. After 3 training iterations, our compact 95M parameter mHuBERT-147 outperforms larger models trained on substantially more data. We rank second and first on the ML-SUPERB 10min and 1h l"},"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":"2406.06371","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T15:32:42Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"03edfe5332c065ce42a92ab72632a5721a62707aa9cdcadb9294829df8b47bd7","abstract_canon_sha256":"0451a43494676273232e59ca78bba81e56c57abd06c94ac2bd55329f79fb8859"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:26.957823Z","signature_b64":"ipnARQClmMtSwWL1tmcOlX+fAG9EqOxTkNYB+IsZN/dxH6ddeUYy3z8alhUrJKAvMOYnD44bAh1s8R5BQauTDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"034a91aedf4d1d1105fb0d40a36a060a452904869bb8e76884f6bde141d5ca3e","last_reissued_at":"2026-07-05T09:38:26.957397Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:26.957397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mHuBERT-147: A Compact Multilingual HuBERT Model","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ioan Calapodescu, Laurent Besacier, Marcely Zanon Boito, Nikolaos Lagos, Vivek Iyer","submitted_at":"2024-06-10T15:32:42Z","abstract_excerpt":"We present mHuBERT-147, the first general-purpose massively multilingual HuBERT speech representation model trained on 90K hours of clean, open-license data. To scale up the multi-iteration HuBERT approach, we use faiss-based clustering, achieving 5.2x faster label assignment than the original method. We also apply a new multilingual batching up-sampling strategy, leveraging both language and dataset diversity. After 3 training iterations, our compact 95M parameter mHuBERT-147 outperforms larger models trained on substantially more data. We rank second and first on the ML-SUPERB 10min and 1h l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06371","kind":"arxiv","version":5},"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/2406.06371/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":"2406.06371","created_at":"2026-07-05T09:38:26.957456+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.06371v5","created_at":"2026-07-05T09:38:26.957456+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06371","created_at":"2026-07-05T09:38:26.957456+00:00"},{"alias_kind":"pith_short_12","alias_value":"ANFJDLW7JUOR","created_at":"2026-07-05T09:38:26.957456+00:00"},{"alias_kind":"pith_short_16","alias_value":"ANFJDLW7JUORCBP3","created_at":"2026-07-05T09:38:26.957456+00:00"},{"alias_kind":"pith_short_8","alias_value":"ANFJDLW7","created_at":"2026-07-05T09:38:26.957456+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27973","citing_title":"From Black-Box to Clinical Insight: A Multi-Stage Explainable Framework for Speech-Based Cognitive Impairment Detection","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ","json":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ.json","graph_json":"https://pith.science/api/pith-number/ANFJDLW7JUORCBP3BVAKG2QGBJ/graph.json","events_json":"https://pith.science/api/pith-number/ANFJDLW7JUORCBP3BVAKG2QGBJ/events.json","paper":"https://pith.science/paper/ANFJDLW7"},"agent_actions":{"view_html":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ","download_json":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ.json","view_paper":"https://pith.science/paper/ANFJDLW7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.06371&json=true","fetch_graph":"https://pith.science/api/pith-number/ANFJDLW7JUORCBP3BVAKG2QGBJ/graph.json","fetch_events":"https://pith.science/api/pith-number/ANFJDLW7JUORCBP3BVAKG2QGBJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ/action/storage_attestation","attest_author":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ/action/author_attestation","sign_citation":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ/action/citation_signature","submit_replication":"https://pith.science/pith/ANFJDLW7JUORCBP3BVAKG2QGBJ/action/replication_record"}},"created_at":"2026-07-05T09:38:26.957456+00:00","updated_at":"2026-07-05T09:38:26.957456+00:00"}