{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BVOE37PUZGGUWEKHFMBSJPOS3E","short_pith_number":"pith:BVOE37PU","schema_version":"1.0","canonical_sha256":"0d5c4dfdf4c98d4b11472b0324bdd2d90ab210c2ccab44b58da62bda4685b99d","source":{"kind":"arxiv","id":"2211.05756","version":1},"attestation_state":"computed","paper":{"title":"Massively Multilingual ASR on 70 Languages: Tokenization, Architecture, and Generalization Capabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Andros Tjandra, David Zhang, Duc Le, Michael L. Seltzer, Nayan Singhal, Ozlem Kalinli","submitted_at":"2022-11-10T18:43:42Z","abstract_excerpt":"End-to-end multilingual ASR has become more appealing because of several reasons such as simplifying the training and deployment process and positive performance transfer from high-resource to low-resource languages. However, scaling up the number of languages, total hours, and number of unique tokens is not a trivial task. This paper explores large-scale multilingual ASR models on 70 languages. We inspect two architectures: (1) Shared embedding and output and (2) Multiple embedding and output model. In the shared model experiments, we show the importance of tokenization strategy across differ"},"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":"2211.05756","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-11-10T18:43:42Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"635686e8fc082649042169c8f2c6bcc2ed8788cc61055be75124e47f0b5a2175","abstract_canon_sha256":"212d225c21d028cc593f1d7f49a00402c6e67d0663d2905ab7a3a46550b82c74"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:15:07.390114Z","signature_b64":"w/xawk4HORSD4wsqHKXPqS3F2nWiLUypYmXdyu5g0h4kBl4nVaa0yrE1Eww0W1BntECRpZXmmY/K7Cv/P44TBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d5c4dfdf4c98d4b11472b0324bdd2d90ab210c2ccab44b58da62bda4685b99d","last_reissued_at":"2026-07-05T05:15:07.389729Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:15:07.389729Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Massively Multilingual ASR on 70 Languages: Tokenization, Architecture, and Generalization Capabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Andros Tjandra, David Zhang, Duc Le, Michael L. Seltzer, Nayan Singhal, Ozlem Kalinli","submitted_at":"2022-11-10T18:43:42Z","abstract_excerpt":"End-to-end multilingual ASR has become more appealing because of several reasons such as simplifying the training and deployment process and positive performance transfer from high-resource to low-resource languages. However, scaling up the number of languages, total hours, and number of unique tokens is not a trivial task. This paper explores large-scale multilingual ASR models on 70 languages. We inspect two architectures: (1) Shared embedding and output and (2) Multiple embedding and output model. In the shared model experiments, we show the importance of tokenization strategy across differ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.05756","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/2211.05756/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":"2211.05756","created_at":"2026-07-05T05:15:07.389786+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.05756v1","created_at":"2026-07-05T05:15:07.389786+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.05756","created_at":"2026-07-05T05:15:07.389786+00:00"},{"alias_kind":"pith_short_12","alias_value":"BVOE37PUZGGU","created_at":"2026-07-05T05:15:07.389786+00:00"},{"alias_kind":"pith_short_16","alias_value":"BVOE37PUZGGUWEKH","created_at":"2026-07-05T05:15:07.389786+00:00"},{"alias_kind":"pith_short_8","alias_value":"BVOE37PU","created_at":"2026-07-05T05:15:07.389786+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/BVOE37PUZGGUWEKHFMBSJPOS3E","json":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E.json","graph_json":"https://pith.science/api/pith-number/BVOE37PUZGGUWEKHFMBSJPOS3E/graph.json","events_json":"https://pith.science/api/pith-number/BVOE37PUZGGUWEKHFMBSJPOS3E/events.json","paper":"https://pith.science/paper/BVOE37PU"},"agent_actions":{"view_html":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E","download_json":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E.json","view_paper":"https://pith.science/paper/BVOE37PU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.05756&json=true","fetch_graph":"https://pith.science/api/pith-number/BVOE37PUZGGUWEKHFMBSJPOS3E/graph.json","fetch_events":"https://pith.science/api/pith-number/BVOE37PUZGGUWEKHFMBSJPOS3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E/action/storage_attestation","attest_author":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E/action/author_attestation","sign_citation":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E/action/citation_signature","submit_replication":"https://pith.science/pith/BVOE37PUZGGUWEKHFMBSJPOS3E/action/replication_record"}},"created_at":"2026-07-05T05:15:07.389786+00:00","updated_at":"2026-07-05T05:15:07.389786+00:00"}