{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:FSZFHIM77WYV4CRFRIAVEEM4YR","short_pith_number":"pith:FSZFHIM7","schema_version":"1.0","canonical_sha256":"2cb253a19ffdb15e0a258a0152119cc462a8973511a2fcf0cb9ac08740adca95","source":{"kind":"arxiv","id":"1908.10992","version":1},"attestation_state":"computed","paper":{"title":"Two-Pass End-to-End Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chung-Cheng Chiu, David Rybach, Ian McGraw, Mirk\\'o Visontai, Qiao Liang, Rohit Prabhavalkar, Ruoming Pang, Tara N. Sainath, Trevor Strohman, Wei Li, Yanzhang He, Yonghui Wu","submitted_at":"2019-08-29T00:18:05Z","abstract_excerpt":"The requirements for many applications of state-of-the-art speech recognition systems include not only low word error rate (WER) but also low latency. Specifically, for many use-cases, the system must be able to decode utterances in a streaming fashion and faster than real-time. Recently, a streaming recurrent neural network transducer (RNN-T) end-to-end (E2E) model has shown to be a good candidate for on-device speech recognition, with improved WER and latency metrics compared to conventional on-device models [1]. However, this model still lags behind a large state-of-the-art conventional mod"},"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":"1908.10992","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-29T00:18:05Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"0038fd916584da4bdeb207dda3e7457f9ebbb2332588bb46e1fb3299df3b9ca3","abstract_canon_sha256":"590409ed1abfed29f67e8eba54d7ced22d11503dd5e6908394a6f2aa5104a7f4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:00:29.188337Z","signature_b64":"x3YJFrB+UTeDHPo7ffupFuLSdm/nqR68RYa3XTDKrtg6gYWMZTzK2TKIylPIx5lpGdjqymOLUhG4IYYQ+klGAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cb253a19ffdb15e0a258a0152119cc462a8973511a2fcf0cb9ac08740adca95","last_reissued_at":"2026-07-05T00:00:29.187865Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:00:29.187865Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Two-Pass End-to-End Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chung-Cheng Chiu, David Rybach, Ian McGraw, Mirk\\'o Visontai, Qiao Liang, Rohit Prabhavalkar, Ruoming Pang, Tara N. Sainath, Trevor Strohman, Wei Li, Yanzhang He, Yonghui Wu","submitted_at":"2019-08-29T00:18:05Z","abstract_excerpt":"The requirements for many applications of state-of-the-art speech recognition systems include not only low word error rate (WER) but also low latency. Specifically, for many use-cases, the system must be able to decode utterances in a streaming fashion and faster than real-time. Recently, a streaming recurrent neural network transducer (RNN-T) end-to-end (E2E) model has shown to be a good candidate for on-device speech recognition, with improved WER and latency metrics compared to conventional on-device models [1]. However, this model still lags behind a large state-of-the-art conventional mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.10992","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1908.10992/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":"1908.10992","created_at":"2026-07-05T00:00:29.187921+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.10992v1","created_at":"2026-07-05T00:00:29.187921+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.10992","created_at":"2026-07-05T00:00:29.187921+00:00"},{"alias_kind":"pith_short_12","alias_value":"FSZFHIM77WYV","created_at":"2026-07-05T00:00:29.187921+00:00"},{"alias_kind":"pith_short_16","alias_value":"FSZFHIM77WYV4CRF","created_at":"2026-07-05T00:00:29.187921+00:00"},{"alias_kind":"pith_short_8","alias_value":"FSZFHIM7","created_at":"2026-07-05T00:00:29.187921+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.07285","citing_title":"Non-Intrusive Automatic Speech Recognition Refinement: A Survey","ref_index":189,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR","json":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR.json","graph_json":"https://pith.science/api/pith-number/FSZFHIM77WYV4CRFRIAVEEM4YR/graph.json","events_json":"https://pith.science/api/pith-number/FSZFHIM77WYV4CRFRIAVEEM4YR/events.json","paper":"https://pith.science/paper/FSZFHIM7"},"agent_actions":{"view_html":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR","download_json":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR.json","view_paper":"https://pith.science/paper/FSZFHIM7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.10992&json=true","fetch_graph":"https://pith.science/api/pith-number/FSZFHIM77WYV4CRFRIAVEEM4YR/graph.json","fetch_events":"https://pith.science/api/pith-number/FSZFHIM77WYV4CRFRIAVEEM4YR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR/action/storage_attestation","attest_author":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR/action/author_attestation","sign_citation":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR/action/citation_signature","submit_replication":"https://pith.science/pith/FSZFHIM77WYV4CRFRIAVEEM4YR/action/replication_record"}},"created_at":"2026-07-05T00:00:29.187921+00:00","updated_at":"2026-07-05T00:00:29.187921+00:00"}