{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:O7VRXEAZGLQ4NAXMOBBHMAGGD3","short_pith_number":"pith:O7VRXEAZ","schema_version":"1.0","canonical_sha256":"77eb1b901932e1c682ec70427600c61ed09e85be8ae5caf8d0462fa033c9976d","source":{"kind":"arxiv","id":"2111.03945","version":3},"attestation_state":"computed","paper":{"title":"Towards Building ASR Systems for the Next Billion Users","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abhigyan Raman, Anoop Kunchukuttan, Gowtham Ramesh, Kaushal Santosh Bhogale, Mitesh M. Khapra, Pratyush Kumar, Sumanth Doddapaneni, Tahir Javed","submitted_at":"2021-11-06T19:34:33Z","abstract_excerpt":"Recent methods in speech and language technology pretrain very LARGE models which are fine-tuned for specific tasks. However, the benefits of such LARGE models are often limited to a few resource rich languages of the world. In this work, we make multiple contributions towards building ASR systems for low resource languages from the Indian subcontinent. First, we curate 17,000 hours of raw speech data for 40 Indian languages from a wide variety of domains including education, news, technology, and finance. Second, using this raw speech data we pretrain several variants of wav2vec style models "},"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":"2111.03945","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-06T19:34:33Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"0ddd7e82b84301191aa669da4273f2b3fb808cd500afb4a471b4e89245b9a038","abstract_canon_sha256":"09e1416bc7fbe52cc8b1c2e91cd8d8c8760d7a95ec4360681b07f3b4133937fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:19:51.351100Z","signature_b64":"yVEPDQ+cH6vJxi/ATVCmpvto/wLlVdXcqjRPhB/fBF7qUZrHudjxYtFJsYvhcv3hpilzY1b2QEH4ShEE9LVDBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77eb1b901932e1c682ec70427600c61ed09e85be8ae5caf8d0462fa033c9976d","last_reissued_at":"2026-07-05T05:19:51.350623Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:19:51.350623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Building ASR Systems for the Next Billion Users","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Abhigyan Raman, Anoop Kunchukuttan, Gowtham Ramesh, Kaushal Santosh Bhogale, Mitesh M. Khapra, Pratyush Kumar, Sumanth Doddapaneni, Tahir Javed","submitted_at":"2021-11-06T19:34:33Z","abstract_excerpt":"Recent methods in speech and language technology pretrain very LARGE models which are fine-tuned for specific tasks. However, the benefits of such LARGE models are often limited to a few resource rich languages of the world. In this work, we make multiple contributions towards building ASR systems for low resource languages from the Indian subcontinent. First, we curate 17,000 hours of raw speech data for 40 Indian languages from a wide variety of domains including education, news, technology, and finance. Second, using this raw speech data we pretrain several variants of wav2vec style models "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.03945","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/2111.03945/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":"2111.03945","created_at":"2026-07-05T05:19:51.350681+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.03945v3","created_at":"2026-07-05T05:19:51.350681+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.03945","created_at":"2026-07-05T05:19:51.350681+00:00"},{"alias_kind":"pith_short_12","alias_value":"O7VRXEAZGLQ4","created_at":"2026-07-05T05:19:51.350681+00:00"},{"alias_kind":"pith_short_16","alias_value":"O7VRXEAZGLQ4NAXM","created_at":"2026-07-05T05:19:51.350681+00:00"},{"alias_kind":"pith_short_8","alias_value":"O7VRXEAZ","created_at":"2026-07-05T05:19:51.350681+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.19797","citing_title":"Enhancing ASR Performance in the Medical Domain for Dravidian Languages","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3","json":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3.json","graph_json":"https://pith.science/api/pith-number/O7VRXEAZGLQ4NAXMOBBHMAGGD3/graph.json","events_json":"https://pith.science/api/pith-number/O7VRXEAZGLQ4NAXMOBBHMAGGD3/events.json","paper":"https://pith.science/paper/O7VRXEAZ"},"agent_actions":{"view_html":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3","download_json":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3.json","view_paper":"https://pith.science/paper/O7VRXEAZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.03945&json=true","fetch_graph":"https://pith.science/api/pith-number/O7VRXEAZGLQ4NAXMOBBHMAGGD3/graph.json","fetch_events":"https://pith.science/api/pith-number/O7VRXEAZGLQ4NAXMOBBHMAGGD3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3/action/storage_attestation","attest_author":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3/action/author_attestation","sign_citation":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3/action/citation_signature","submit_replication":"https://pith.science/pith/O7VRXEAZGLQ4NAXMOBBHMAGGD3/action/replication_record"}},"created_at":"2026-07-05T05:19:51.350681+00:00","updated_at":"2026-07-05T05:19:51.350681+00:00"}