{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:KJKTX77O72JFZ7XYNJE5XQPTAX","short_pith_number":"pith:KJKTX77O","schema_version":"1.0","canonical_sha256":"52553bffeefe925cfef86a49dbc1f305c69170da55e35d03ad1411fd262a8286","source":{"kind":"arxiv","id":"2006.13979","version":2},"attestation_state":"computed","paper":{"title":"Unsupervised Cross-lingual Representation Learning for Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alexei Baevski, Alexis Conneau, Michael Auli, Ronan Collobert","submitted_at":"2020-06-24T18:25:05Z","abstract_excerpt":"This paper presents XLSR which learns cross-lingual speech representations by pretraining a single model from the raw waveform of speech in multiple languages. We build on wav2vec 2.0 which is trained by solving a contrastive task over masked latent speech representations and jointly learns a quantization of the latents shared across languages. The resulting model is fine-tuned on labeled data and experiments show that cross-lingual pretraining significantly outperforms monolingual pretraining. On the CommonVoice benchmark, XLSR shows a relative phoneme error rate reduction of 72% compared to "},"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":"2006.13979","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-24T18:25:05Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"34b8c3b8615d4c69ace55bcfeb86741d431b554680fad89f5ed5314e5dd2a527","abstract_canon_sha256":"8344f1951c320d3bd46394fafe7cbae030b63b3c5da93b714179565b52245380"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:59:56.119490Z","signature_b64":"MAqt2SZ+Bl6pyn6AC4KsqpidP/cBBhpm2f3tMMHUwRxnVg+1+6elO+UQxW43e0aaRq0wH6MRHQLlw7K59EV5DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52553bffeefe925cfef86a49dbc1f305c69170da55e35d03ad1411fd262a8286","last_reissued_at":"2026-07-05T01:59:56.118988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:59:56.118988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised Cross-lingual Representation Learning for Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alexei Baevski, Alexis Conneau, Michael Auli, Ronan Collobert","submitted_at":"2020-06-24T18:25:05Z","abstract_excerpt":"This paper presents XLSR which learns cross-lingual speech representations by pretraining a single model from the raw waveform of speech in multiple languages. We build on wav2vec 2.0 which is trained by solving a contrastive task over masked latent speech representations and jointly learns a quantization of the latents shared across languages. The resulting model is fine-tuned on labeled data and experiments show that cross-lingual pretraining significantly outperforms monolingual pretraining. On the CommonVoice benchmark, XLSR shows a relative phoneme error rate reduction of 72% compared to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.13979","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/2006.13979/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":"2006.13979","created_at":"2026-07-05T01:59:56.119045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.13979v2","created_at":"2026-07-05T01:59:56.119045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.13979","created_at":"2026-07-05T01:59:56.119045+00:00"},{"alias_kind":"pith_short_12","alias_value":"KJKTX77O72JF","created_at":"2026-07-05T01:59:56.119045+00:00"},{"alias_kind":"pith_short_16","alias_value":"KJKTX77O72JFZ7XY","created_at":"2026-07-05T01:59:56.119045+00:00"},{"alias_kind":"pith_short_8","alias_value":"KJKTX77O","created_at":"2026-07-05T01:59:56.119045+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21990","citing_title":"Adding Robust Code-Switching Capabilities to High Performance Multilingual ASR","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11542","citing_title":"Pretrained self-supervised speech models can recognize unseen consonants","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09345","citing_title":"A study on the impact of region specific data on the performance of Indic ASR","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09535","citing_title":"Overcoming Decoder Inconsistencies in Whisper for Dravidian and Low-Resource Languages","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2510.10785","citing_title":"FAC-FACodec: Controllable Zero-Shot Foreign Accent Conversion with Factorized Speech Codec","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03298","citing_title":"ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01101","citing_title":"Virtual Speech Therapist: A Clinician-in-the-Loop AI Speech Therapy Agent for Personalized and Supervised Therapy","ref_index":47,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX","json":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX.json","graph_json":"https://pith.science/api/pith-number/KJKTX77O72JFZ7XYNJE5XQPTAX/graph.json","events_json":"https://pith.science/api/pith-number/KJKTX77O72JFZ7XYNJE5XQPTAX/events.json","paper":"https://pith.science/paper/KJKTX77O"},"agent_actions":{"view_html":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX","download_json":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX.json","view_paper":"https://pith.science/paper/KJKTX77O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.13979&json=true","fetch_graph":"https://pith.science/api/pith-number/KJKTX77O72JFZ7XYNJE5XQPTAX/graph.json","fetch_events":"https://pith.science/api/pith-number/KJKTX77O72JFZ7XYNJE5XQPTAX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX/action/storage_attestation","attest_author":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX/action/author_attestation","sign_citation":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX/action/citation_signature","submit_replication":"https://pith.science/pith/KJKTX77O72JFZ7XYNJE5XQPTAX/action/replication_record"}},"created_at":"2026-07-05T01:59:56.119045+00:00","updated_at":"2026-07-05T01:59:56.119045+00:00"}