{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HXWMX2MY76VFBLLF23A3OC5MMG","short_pith_number":"pith:HXWMX2MY","schema_version":"1.0","canonical_sha256":"3deccbe998ffaa50ad65d6c1b70bac61984e8451527e012215ce6ece9cf0b8a7","source":{"kind":"arxiv","id":"2406.11546","version":2},"attestation_state":"computed","paper":{"title":"GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Bo Yang, Guoguo Chen, Jianheng Zhuo, Jinpeng Li, Kai Yu, Ke Li, Mingyu Cui, Shuai Fan, Wei-Qiang Zhang, Xie Chen, Xunying Liu, Yexing Du, Yifan Yang, Zheshu Song, Ziyang Ma, Ziyuan Wang","submitted_at":"2024-06-17T13:44:20Z","abstract_excerpt":"The evolution of speech technology has been spurred by the rapid increase in dataset sizes. Traditional speech models generally depend on a large amount of labeled training data, which is scarce for low-resource languages. This paper presents GigaSpeech 2, a large-scale, multi-domain, multilingual speech recognition corpus. It is designed for low-resource languages and does not rely on paired speech and text data. GigaSpeech 2 comprises about 30,000 hours of automatically transcribed speech, including Thai, Indonesian, and Vietnamese, gathered from unlabeled YouTube videos. We also introduce a"},"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":"2406.11546","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2024-06-17T13:44:20Z","cross_cats_sorted":["cs.CL","cs.SD"],"title_canon_sha256":"b15bfa94c2b1ec4a1944bf578350284d272a3b7832a161186fea099fbb00e9cb","abstract_canon_sha256":"0c67afb709983a764f09fc3bfc5eaaafd0ee477019e8e2e56f7c37cb9d673c0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:59.604540Z","signature_b64":"WzIyB4gBdEVNV44oiw4m1aZInO1r6+CRqeiCcD50J4LtN+6Zd7TV9TXXAXiQvG6FyUVzOCYB/QFv0e3S8VMcDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3deccbe998ffaa50ad65d6c1b70bac61984e8451527e012215ce6ece9cf0b8a7","last_reissued_at":"2026-07-05T11:09:59.604002Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:59.604002Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Bo Yang, Guoguo Chen, Jianheng Zhuo, Jinpeng Li, Kai Yu, Ke Li, Mingyu Cui, Shuai Fan, Wei-Qiang Zhang, Xie Chen, Xunying Liu, Yexing Du, Yifan Yang, Zheshu Song, Ziyang Ma, Ziyuan Wang","submitted_at":"2024-06-17T13:44:20Z","abstract_excerpt":"The evolution of speech technology has been spurred by the rapid increase in dataset sizes. Traditional speech models generally depend on a large amount of labeled training data, which is scarce for low-resource languages. This paper presents GigaSpeech 2, a large-scale, multi-domain, multilingual speech recognition corpus. It is designed for low-resource languages and does not rely on paired speech and text data. GigaSpeech 2 comprises about 30,000 hours of automatically transcribed speech, including Thai, Indonesian, and Vietnamese, gathered from unlabeled YouTube videos. We also introduce a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11546","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/2406.11546/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":"2406.11546","created_at":"2026-07-05T11:09:59.604065+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.11546v2","created_at":"2026-07-05T11:09:59.604065+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11546","created_at":"2026-07-05T11:09:59.604065+00:00"},{"alias_kind":"pith_short_12","alias_value":"HXWMX2MY76VF","created_at":"2026-07-05T11:09:59.604065+00:00"},{"alias_kind":"pith_short_16","alias_value":"HXWMX2MY76VFBLLF","created_at":"2026-07-05T11:09:59.604065+00:00"},{"alias_kind":"pith_short_8","alias_value":"HXWMX2MY","created_at":"2026-07-05T11:09:59.604065+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.14804","citing_title":"Towards Building Speech Large Language Models for Multitask Understanding in Low-Resource Languages","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08962","citing_title":"MegaScale-Omni: A Hyper-Scale, Workload-Resilient System for MultiModal LLM Training in Production","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG","json":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG.json","graph_json":"https://pith.science/api/pith-number/HXWMX2MY76VFBLLF23A3OC5MMG/graph.json","events_json":"https://pith.science/api/pith-number/HXWMX2MY76VFBLLF23A3OC5MMG/events.json","paper":"https://pith.science/paper/HXWMX2MY"},"agent_actions":{"view_html":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG","download_json":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG.json","view_paper":"https://pith.science/paper/HXWMX2MY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.11546&json=true","fetch_graph":"https://pith.science/api/pith-number/HXWMX2MY76VFBLLF23A3OC5MMG/graph.json","fetch_events":"https://pith.science/api/pith-number/HXWMX2MY76VFBLLF23A3OC5MMG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG/action/storage_attestation","attest_author":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG/action/author_attestation","sign_citation":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG/action/citation_signature","submit_replication":"https://pith.science/pith/HXWMX2MY76VFBLLF23A3OC5MMG/action/replication_record"}},"created_at":"2026-07-05T11:09:59.604065+00:00","updated_at":"2026-07-05T11:09:59.604065+00:00"}