{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6EAPG53PXBESNWDBN7ENKHX2IG","short_pith_number":"pith:6EAPG53P","schema_version":"1.0","canonical_sha256":"f100f3776fb84926d8616fc8d51efa41b2b3f69987da6a0a67614b919493abad","source":{"kind":"arxiv","id":"2305.10615","version":3},"attestation_state":"computed","paper":{"title":"ML-SUPERB: Multilingual Speech Universal PERformance Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Abdelrahman Mohamed, Dan Berrebbi, En-Pei Hu, Ho-Lam Chung, Hung-yi Lee, Jiatong Shi, Shang-Wen Li, Shinji Watanabe, Wei Ping Huang, William Chen, Xuankai Chang","submitted_at":"2023-05-18T00:01:27Z","abstract_excerpt":"Speech processing Universal PERformance Benchmark (SUPERB) is a leaderboard to benchmark the performance of Self-Supervised Learning (SSL) models on various speech processing tasks. However, SUPERB largely considers English speech in its evaluation. This paper presents multilingual SUPERB (ML-SUPERB), covering 143 languages (ranging from high-resource to endangered), and considering both automatic speech recognition and language identification. Following the concept of SUPERB, ML-SUPERB utilizes frozen SSL features and employs a simple framework for multilingual tasks by learning a shallow dow"},"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":"2305.10615","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2023-05-18T00:01:27Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"dfb50610efc8e8d54e48317a59036eac99eddaa3b1d2435c6648c52b884eaa65","abstract_canon_sha256":"d9dd39a28b4861021cac6ac41c5a6df5cf4a210bda874ce89a1e63eef1a34ef3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:31.877960Z","signature_b64":"689vCdViqIXzteRCgzamLX9WGuEdij9/mazNVSRGv2GjDicW1qGbvZG91wwWXlX5zlkrhCFzZd9NNs97DTKxDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f100f3776fb84926d8616fc8d51efa41b2b3f69987da6a0a67614b919493abad","last_reissued_at":"2026-07-05T10:18:31.877572Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:31.877572Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ML-SUPERB: Multilingual Speech Universal PERformance Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Abdelrahman Mohamed, Dan Berrebbi, En-Pei Hu, Ho-Lam Chung, Hung-yi Lee, Jiatong Shi, Shang-Wen Li, Shinji Watanabe, Wei Ping Huang, William Chen, Xuankai Chang","submitted_at":"2023-05-18T00:01:27Z","abstract_excerpt":"Speech processing Universal PERformance Benchmark (SUPERB) is a leaderboard to benchmark the performance of Self-Supervised Learning (SSL) models on various speech processing tasks. However, SUPERB largely considers English speech in its evaluation. This paper presents multilingual SUPERB (ML-SUPERB), covering 143 languages (ranging from high-resource to endangered), and considering both automatic speech recognition and language identification. Following the concept of SUPERB, ML-SUPERB utilizes frozen SSL features and employs a simple framework for multilingual tasks by learning a shallow dow"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.10615","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/2305.10615/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":"2305.10615","created_at":"2026-07-05T10:18:31.877621+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.10615v3","created_at":"2026-07-05T10:18:31.877621+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10615","created_at":"2026-07-05T10:18:31.877621+00:00"},{"alias_kind":"pith_short_12","alias_value":"6EAPG53PXBES","created_at":"2026-07-05T10:18:31.877621+00:00"},{"alias_kind":"pith_short_16","alias_value":"6EAPG53PXBESNWDB","created_at":"2026-07-05T10:18:31.877621+00:00"},{"alias_kind":"pith_short_8","alias_value":"6EAPG53P","created_at":"2026-07-05T10:18:31.877621+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.01482","citing_title":"A SUPERB-Style Benchmark of Self-Supervised Speech Models for Audio Deepfake Detection","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10494","citing_title":"Multi-layer attentive probing improves transfer of audio representations for bioacoustics","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10736","citing_title":"BlasBench: An Open Benchmark for Irish Speech Recognition","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01381","citing_title":"A framework for analyzing concept representations in neural models","ref_index":205,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG","json":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG.json","graph_json":"https://pith.science/api/pith-number/6EAPG53PXBESNWDBN7ENKHX2IG/graph.json","events_json":"https://pith.science/api/pith-number/6EAPG53PXBESNWDBN7ENKHX2IG/events.json","paper":"https://pith.science/paper/6EAPG53P"},"agent_actions":{"view_html":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG","download_json":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG.json","view_paper":"https://pith.science/paper/6EAPG53P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.10615&json=true","fetch_graph":"https://pith.science/api/pith-number/6EAPG53PXBESNWDBN7ENKHX2IG/graph.json","fetch_events":"https://pith.science/api/pith-number/6EAPG53PXBESNWDBN7ENKHX2IG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG/action/storage_attestation","attest_author":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG/action/author_attestation","sign_citation":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG/action/citation_signature","submit_replication":"https://pith.science/pith/6EAPG53PXBESNWDBN7ENKHX2IG/action/replication_record"}},"created_at":"2026-07-05T10:18:31.877621+00:00","updated_at":"2026-07-05T10:18:31.877621+00:00"}