{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:CPLGYHLQOSB5J5WTA4LJNF54PZ","short_pith_number":"pith:CPLGYHLQ","schema_version":"1.0","canonical_sha256":"13d66c1d707483d4f6d307169697bc7e4d07ce79b328304fca543c6adf6b23b6","source":{"kind":"arxiv","id":"1907.01372","version":1},"attestation_state":"computed","paper":{"title":"Improving Performance of End-to-End ASR on Numeric Sequences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Cal Peyser, Hao Zhang, Tara N. Sainath, Zelin Wu","submitted_at":"2019-07-01T14:21:09Z","abstract_excerpt":"Recognizing written domain numeric utterances (e.g. I need $1.25.) can be challenging for ASR systems, particularly when numeric sequences are not seen during training. This out-of-vocabulary (OOV) issue is addressed in conventional ASR systems by training part of the model on spoken domain utterances (e.g. I need one dollar and twenty five cents.), for which numeric sequences are composed of in-vocabulary numbers, and then using an FST verbalizer to denormalize the result. Unfortunately, conventional ASR models are not suitable for the low memory setting of on-device speech recognition. E2E m"},"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":"1907.01372","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2019-07-01T14:21:09Z","cross_cats_sorted":["cs.LG","cs.SD","stat.ML"],"title_canon_sha256":"12bf897a90bfdbd14b2845502b32114fdab7fea499deb79e8760a58ca4da6dd5","abstract_canon_sha256":"40cfed1b62eb5249e6cfdb055649c278f66b295e7834e4e428f7af63749b4e44"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:41:40.573226Z","signature_b64":"6ace1CEWz0//mFzdfB7bkMgvXCZbp7pJR03f3ryI83VOfk1RrlZkP7nb9SGIeE1pb1KQKGMVkthVdpKXW6L8DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"13d66c1d707483d4f6d307169697bc7e4d07ce79b328304fca543c6adf6b23b6","last_reissued_at":"2026-05-17T23:41:40.572615Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:41:40.572615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Performance of End-to-End ASR on Numeric Sequences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Cal Peyser, Hao Zhang, Tara N. Sainath, Zelin Wu","submitted_at":"2019-07-01T14:21:09Z","abstract_excerpt":"Recognizing written domain numeric utterances (e.g. I need $1.25.) can be challenging for ASR systems, particularly when numeric sequences are not seen during training. This out-of-vocabulary (OOV) issue is addressed in conventional ASR systems by training part of the model on spoken domain utterances (e.g. I need one dollar and twenty five cents.), for which numeric sequences are composed of in-vocabulary numbers, and then using an FST verbalizer to denormalize the result. Unfortunately, conventional ASR models are not suitable for the low memory setting of on-device speech recognition. E2E m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.01372","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1907.01372","created_at":"2026-05-17T23:41:40.572704+00:00"},{"alias_kind":"arxiv_version","alias_value":"1907.01372v1","created_at":"2026-05-17T23:41:40.572704+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.01372","created_at":"2026-05-17T23:41:40.572704+00:00"},{"alias_kind":"pith_short_12","alias_value":"CPLGYHLQOSB5","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"CPLGYHLQOSB5J5WT","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"CPLGYHLQ","created_at":"2026-05-18T12:33:15.570797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.01372","citing_title":"Improving Performance of End-to-End ASR on Numeric Sequences","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ","json":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ.json","graph_json":"https://pith.science/api/pith-number/CPLGYHLQOSB5J5WTA4LJNF54PZ/graph.json","events_json":"https://pith.science/api/pith-number/CPLGYHLQOSB5J5WTA4LJNF54PZ/events.json","paper":"https://pith.science/paper/CPLGYHLQ"},"agent_actions":{"view_html":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ","download_json":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ.json","view_paper":"https://pith.science/paper/CPLGYHLQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1907.01372&json=true","fetch_graph":"https://pith.science/api/pith-number/CPLGYHLQOSB5J5WTA4LJNF54PZ/graph.json","fetch_events":"https://pith.science/api/pith-number/CPLGYHLQOSB5J5WTA4LJNF54PZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ/action/storage_attestation","attest_author":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ/action/author_attestation","sign_citation":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ/action/citation_signature","submit_replication":"https://pith.science/pith/CPLGYHLQOSB5J5WTA4LJNF54PZ/action/replication_record"}},"created_at":"2026-05-17T23:41:40.572704+00:00","updated_at":"2026-05-17T23:41:40.572704+00:00"}