{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:IZ3FJAMK5AUVLTHQJBVFNRSYNH","short_pith_number":"pith:IZ3FJAMK","schema_version":"1.0","canonical_sha256":"467654818ae82955ccf0486a56c65869c7397f9accecbc4db8065b83290508df","source":{"kind":"arxiv","id":"2012.07974","version":3},"attestation_state":"computed","paper":{"title":"A review of on-device fully neural end-to-end automatic speech recognition algorithms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Abhinav Garg, Ankur Kumar, Changwoo Han, Chanwoo Kim, Dhananjaya Gowda, Dongsoo Lee, Jiyeon Kim, Sungsoo Kim","submitted_at":"2020-12-14T22:18:08Z","abstract_excerpt":"In this paper, we review various end-to-end automatic speech recognition algorithms and their optimization techniques for on-device applications. Conventional speech recognition systems comprise a large number of discrete components such as an acoustic model, a language model, a pronunciation model, a text-normalizer, an inverse-text normalizer, a decoder based on a Weighted Finite State Transducer (WFST), and so on. To obtain sufficiently high speech recognition accuracy with such conventional speech recognition systems, a very large language model (up to 100 GB) is usually needed. Hence, the"},"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":"2012.07974","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-12-14T22:18:08Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b56582297017d43484ff98688de44bbe33de203f0a71305a85acf997140e2901","abstract_canon_sha256":"bb7198c0f46561b9a58a4739d0bcb26e62066f7bc6d38c33b2c3407dda60300e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:09:16.549416Z","signature_b64":"HTfgogBI50avV4Q8OLPoZ5Fow5GqqLi6ZkjZd6XO5onIXBiMEzgcdISnLUChcwiUzr3M4pF3DQZDQFiET/M6BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"467654818ae82955ccf0486a56c65869c7397f9accecbc4db8065b83290508df","last_reissued_at":"2026-07-05T03:09:16.548962Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:09:16.548962Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A review of on-device fully neural end-to-end automatic speech recognition algorithms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Abhinav Garg, Ankur Kumar, Changwoo Han, Chanwoo Kim, Dhananjaya Gowda, Dongsoo Lee, Jiyeon Kim, Sungsoo Kim","submitted_at":"2020-12-14T22:18:08Z","abstract_excerpt":"In this paper, we review various end-to-end automatic speech recognition algorithms and their optimization techniques for on-device applications. Conventional speech recognition systems comprise a large number of discrete components such as an acoustic model, a language model, a pronunciation model, a text-normalizer, an inverse-text normalizer, a decoder based on a Weighted Finite State Transducer (WFST), and so on. To obtain sufficiently high speech recognition accuracy with such conventional speech recognition systems, a very large language model (up to 100 GB) is usually needed. Hence, the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.07974","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/2012.07974/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":"2012.07974","created_at":"2026-07-05T03:09:16.549024+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.07974v3","created_at":"2026-07-05T03:09:16.549024+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.07974","created_at":"2026-07-05T03:09:16.549024+00:00"},{"alias_kind":"pith_short_12","alias_value":"IZ3FJAMK5AUV","created_at":"2026-07-05T03:09:16.549024+00:00"},{"alias_kind":"pith_short_16","alias_value":"IZ3FJAMK5AUVLTHQ","created_at":"2026-07-05T03:09:16.549024+00:00"},{"alias_kind":"pith_short_8","alias_value":"IZ3FJAMK","created_at":"2026-07-05T03:09:16.549024+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/IZ3FJAMK5AUVLTHQJBVFNRSYNH","json":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH.json","graph_json":"https://pith.science/api/pith-number/IZ3FJAMK5AUVLTHQJBVFNRSYNH/graph.json","events_json":"https://pith.science/api/pith-number/IZ3FJAMK5AUVLTHQJBVFNRSYNH/events.json","paper":"https://pith.science/paper/IZ3FJAMK"},"agent_actions":{"view_html":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH","download_json":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH.json","view_paper":"https://pith.science/paper/IZ3FJAMK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.07974&json=true","fetch_graph":"https://pith.science/api/pith-number/IZ3FJAMK5AUVLTHQJBVFNRSYNH/graph.json","fetch_events":"https://pith.science/api/pith-number/IZ3FJAMK5AUVLTHQJBVFNRSYNH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH/action/storage_attestation","attest_author":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH/action/author_attestation","sign_citation":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH/action/citation_signature","submit_replication":"https://pith.science/pith/IZ3FJAMK5AUVLTHQJBVFNRSYNH/action/replication_record"}},"created_at":"2026-07-05T03:09:16.549024+00:00","updated_at":"2026-07-05T03:09:16.549024+00:00"}