{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:WFEX3WLZGGZXOQRVPB26FE764F","short_pith_number":"pith:WFEX3WLZ","schema_version":"1.0","canonical_sha256":"b1497dd97931b37742357875e293fee14c81d39eea26a02cb0d77c8ded94eba9","source":{"kind":"arxiv","id":"1906.04323","version":2},"attestation_state":"computed","paper":{"title":"Word-level Speech Recognition with a Letter to Word Encoder","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Awni Hannun, Gabriel Synnaeve, Ronan Collobert","submitted_at":"2019-06-10T23:31:56Z","abstract_excerpt":"We propose a direct-to-word sequence model which uses a word network to learn word embeddings from letters. The word network can be integrated seamlessly with arbitrary sequence models including Connectionist Temporal Classification and encoder-decoder models with attention. We show our direct-to-word model can achieve word error rate gains over sub-word level models for speech recognition. We also show that our direct-to-word approach retains the ability to predict words not seen at training time without any retraining. Finally, we demonstrate that a word-level model can use a larger stride t"},"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":"1906.04323","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-10T23:31:56Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"c85fba5050cb0e33167e4db5ced95fede9fde239df3740844ed3e70093ee0a40","abstract_canon_sha256":"abbf8429a88a5f035e912b970a092449dea215b6d8ea220dc58a651b09e5b88b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:19:14.335299Z","signature_b64":"mKEO+DOtxi2wdA2V9UQ6MhzeBbkqVALxdc2uHCPGBkE8EKioWzpDihdwA2k/fUHixCYIZZY2l0dcGKEmOpQJAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b1497dd97931b37742357875e293fee14c81d39eea26a02cb0d77c8ded94eba9","last_reissued_at":"2026-07-05T01:19:14.334767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:19:14.334767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Word-level Speech Recognition with a Letter to Word Encoder","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Awni Hannun, Gabriel Synnaeve, Ronan Collobert","submitted_at":"2019-06-10T23:31:56Z","abstract_excerpt":"We propose a direct-to-word sequence model which uses a word network to learn word embeddings from letters. The word network can be integrated seamlessly with arbitrary sequence models including Connectionist Temporal Classification and encoder-decoder models with attention. We show our direct-to-word model can achieve word error rate gains over sub-word level models for speech recognition. We also show that our direct-to-word approach retains the ability to predict words not seen at training time without any retraining. Finally, we demonstrate that a word-level model can use a larger stride t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.04323","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/1906.04323/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":"1906.04323","created_at":"2026-07-05T01:19:14.334826+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.04323v2","created_at":"2026-07-05T01:19:14.334826+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.04323","created_at":"2026-07-05T01:19:14.334826+00:00"},{"alias_kind":"pith_short_12","alias_value":"WFEX3WLZGGZX","created_at":"2026-07-05T01:19:14.334826+00:00"},{"alias_kind":"pith_short_16","alias_value":"WFEX3WLZGGZXOQRV","created_at":"2026-07-05T01:19:14.334826+00:00"},{"alias_kind":"pith_short_8","alias_value":"WFEX3WLZ","created_at":"2026-07-05T01:19:14.334826+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/WFEX3WLZGGZXOQRVPB26FE764F","json":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F.json","graph_json":"https://pith.science/api/pith-number/WFEX3WLZGGZXOQRVPB26FE764F/graph.json","events_json":"https://pith.science/api/pith-number/WFEX3WLZGGZXOQRVPB26FE764F/events.json","paper":"https://pith.science/paper/WFEX3WLZ"},"agent_actions":{"view_html":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F","download_json":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F.json","view_paper":"https://pith.science/paper/WFEX3WLZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.04323&json=true","fetch_graph":"https://pith.science/api/pith-number/WFEX3WLZGGZXOQRVPB26FE764F/graph.json","fetch_events":"https://pith.science/api/pith-number/WFEX3WLZGGZXOQRVPB26FE764F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F/action/storage_attestation","attest_author":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F/action/author_attestation","sign_citation":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F/action/citation_signature","submit_replication":"https://pith.science/pith/WFEX3WLZGGZXOQRVPB26FE764F/action/replication_record"}},"created_at":"2026-07-05T01:19:14.334826+00:00","updated_at":"2026-07-05T01:19:14.334826+00:00"}