{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:WECHKIVZILCM2NDMKYNAZYRWQC","short_pith_number":"pith:WECHKIVZ","schema_version":"1.0","canonical_sha256":"b1047522b942c4cd346c561a0ce23680b20730215ccdef1f7597949a98af3ed4","source":{"kind":"arxiv","id":"2110.03370","version":5},"attestation_state":"computed","paper":{"title":"WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Binbin Zhang, Chao Yang, Chenchen Zeng, Di Wu, Hang Lv, Hui Bu, Lei Xie, Pengcheng Guo, Qijie Shao, Xiaoyu Chen, Xin Xu, Zhendong Peng","submitted_at":"2021-10-07T12:05:29Z","abstract_excerpt":"In this paper, we present WenetSpeech, a multi-domain Mandarin corpus consisting of 10000+ hours high-quality labeled speech, 2400+ hours weakly labeled speech, and about 10000 hours unlabeled speech, with 22400+ hours in total. We collect the data from YouTube and Podcast, which covers a variety of speaking styles, scenarios, domains, topics, and noisy conditions. An optical character recognition (OCR) based method is introduced to generate the audio/text segmentation candidates for the YouTube data on its corresponding video captions, while a high-quality ASR transcription system is used 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":"2110.03370","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2021-10-07T12:05:29Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"fa99ec6d25c195b5d31fa6f600c5c49caf9b8695e594166ac283b7947ada4bd5","abstract_canon_sha256":"3bdfc17293d3f52349169bfa8732825e5cd51ee1f18964e520d6c6b23efd6694"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:59:24.891083Z","signature_b64":"hkReyBcUZPM8kqOceOQIRkVsHEzI0X53aOjqLVSLzbSVk8TftkKlaro33IChTyxYZnp4FdX+okU/ToGjTWuqCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b1047522b942c4cd346c561a0ce23680b20730215ccdef1f7597949a98af3ed4","last_reissued_at":"2026-07-05T03:59:24.890604Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:59:24.890604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Binbin Zhang, Chao Yang, Chenchen Zeng, Di Wu, Hang Lv, Hui Bu, Lei Xie, Pengcheng Guo, Qijie Shao, Xiaoyu Chen, Xin Xu, Zhendong Peng","submitted_at":"2021-10-07T12:05:29Z","abstract_excerpt":"In this paper, we present WenetSpeech, a multi-domain Mandarin corpus consisting of 10000+ hours high-quality labeled speech, 2400+ hours weakly labeled speech, and about 10000 hours unlabeled speech, with 22400+ hours in total. We collect the data from YouTube and Podcast, which covers a variety of speaking styles, scenarios, domains, topics, and noisy conditions. An optical character recognition (OCR) based method is introduced to generate the audio/text segmentation candidates for the YouTube data on its corresponding video captions, while a high-quality ASR transcription system is used to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03370","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/2110.03370/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":"2110.03370","created_at":"2026-07-05T03:59:24.890663+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.03370v5","created_at":"2026-07-05T03:59:24.890663+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03370","created_at":"2026-07-05T03:59:24.890663+00:00"},{"alias_kind":"pith_short_12","alias_value":"WECHKIVZILCM","created_at":"2026-07-05T03:59:24.890663+00:00"},{"alias_kind":"pith_short_16","alias_value":"WECHKIVZILCM2NDM","created_at":"2026-07-05T03:59:24.890663+00:00"},{"alias_kind":"pith_short_8","alias_value":"WECHKIVZ","created_at":"2026-07-05T03:59:24.890663+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.28139","citing_title":"Data-Efficient On-Policy Distillation for Automatic Speech Recognition","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2508.07285","citing_title":"Non-Intrusive Automatic Speech Recognition Refinement: A Survey","ref_index":136,"is_internal_anchor":false},{"citing_arxiv_id":"2507.16632","citing_title":"Step-Audio 2 Technical Report","ref_index":80,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC","json":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC.json","graph_json":"https://pith.science/api/pith-number/WECHKIVZILCM2NDMKYNAZYRWQC/graph.json","events_json":"https://pith.science/api/pith-number/WECHKIVZILCM2NDMKYNAZYRWQC/events.json","paper":"https://pith.science/paper/WECHKIVZ"},"agent_actions":{"view_html":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC","download_json":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC.json","view_paper":"https://pith.science/paper/WECHKIVZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.03370&json=true","fetch_graph":"https://pith.science/api/pith-number/WECHKIVZILCM2NDMKYNAZYRWQC/graph.json","fetch_events":"https://pith.science/api/pith-number/WECHKIVZILCM2NDMKYNAZYRWQC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC/action/storage_attestation","attest_author":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC/action/author_attestation","sign_citation":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC/action/citation_signature","submit_replication":"https://pith.science/pith/WECHKIVZILCM2NDMKYNAZYRWQC/action/replication_record"}},"created_at":"2026-07-05T03:59:24.890663+00:00","updated_at":"2026-07-05T03:59:24.890663+00:00"}