{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:7BPB43YHCIQE35U5TOLRPJFJAM","short_pith_number":"pith:7BPB43YH","schema_version":"1.0","canonical_sha256":"f85e1e6f0712204df69d9b9717a4a9032961ae88da5fb6e67854e413bff8eb09","source":{"kind":"arxiv","id":"2011.03072","version":1},"attestation_state":"computed","paper":{"title":"Alignment Restricted Streaming Recurrent Neural Network Transducer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ching-Feng Yeh, Christian Fuegen, Duc Le, Gil Keren, Hang Su, Jay Mahadeokar, Michael L. Seltzer, Thong Le, Yuan Shangguan","submitted_at":"2020-11-05T19:38:54Z","abstract_excerpt":"There is a growing interest in the speech community in developing Recurrent Neural Network Transducer (RNN-T) models for automatic speech recognition (ASR) applications. RNN-T is trained with a loss function that does not enforce temporal alignment of the training transcripts and audio. As a result, RNN-T models built with uni-directional long short term memory (LSTM) encoders tend to wait for longer spans of input audio, before streaming already decoded ASR tokens. In this work, we propose a modification to the RNN-T loss function and develop Alignment Restricted RNN-T (Ar-RNN-T) models, whic"},"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":"2011.03072","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-05T19:38:54Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"09dd842d7c3a6a5e0e9a675d4e784a5c5928d502aa67dfbca2fb3b5418223316","abstract_canon_sha256":"3d6ad2d046ad45ee07c9d1e47f1da6603755fe0a208df32936e3617395875b08"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:53:00.052709Z","signature_b64":"n/cUpi9w6jzR7SBEtVgepCbjyl/M/AKO5D46tjHtVvrL6KYZornkiVLjARqH5ZOCp7k1TxNeYyGyHs52jEawAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f85e1e6f0712204df69d9b9717a4a9032961ae88da5fb6e67854e413bff8eb09","last_reissued_at":"2026-07-05T01:53:00.052260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:53:00.052260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Alignment Restricted Streaming Recurrent Neural Network Transducer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ching-Feng Yeh, Christian Fuegen, Duc Le, Gil Keren, Hang Su, Jay Mahadeokar, Michael L. Seltzer, Thong Le, Yuan Shangguan","submitted_at":"2020-11-05T19:38:54Z","abstract_excerpt":"There is a growing interest in the speech community in developing Recurrent Neural Network Transducer (RNN-T) models for automatic speech recognition (ASR) applications. RNN-T is trained with a loss function that does not enforce temporal alignment of the training transcripts and audio. As a result, RNN-T models built with uni-directional long short term memory (LSTM) encoders tend to wait for longer spans of input audio, before streaming already decoded ASR tokens. In this work, we propose a modification to the RNN-T loss function and develop Alignment Restricted RNN-T (Ar-RNN-T) models, whic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.03072","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/2011.03072/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":"2011.03072","created_at":"2026-07-05T01:53:00.052326+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.03072v1","created_at":"2026-07-05T01:53:00.052326+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.03072","created_at":"2026-07-05T01:53:00.052326+00:00"},{"alias_kind":"pith_short_12","alias_value":"7BPB43YHCIQE","created_at":"2026-07-05T01:53:00.052326+00:00"},{"alias_kind":"pith_short_16","alias_value":"7BPB43YHCIQE35U5","created_at":"2026-07-05T01:53:00.052326+00:00"},{"alias_kind":"pith_short_8","alias_value":"7BPB43YH","created_at":"2026-07-05T01:53:00.052326+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/7BPB43YHCIQE35U5TOLRPJFJAM","json":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM.json","graph_json":"https://pith.science/api/pith-number/7BPB43YHCIQE35U5TOLRPJFJAM/graph.json","events_json":"https://pith.science/api/pith-number/7BPB43YHCIQE35U5TOLRPJFJAM/events.json","paper":"https://pith.science/paper/7BPB43YH"},"agent_actions":{"view_html":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM","download_json":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM.json","view_paper":"https://pith.science/paper/7BPB43YH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.03072&json=true","fetch_graph":"https://pith.science/api/pith-number/7BPB43YHCIQE35U5TOLRPJFJAM/graph.json","fetch_events":"https://pith.science/api/pith-number/7BPB43YHCIQE35U5TOLRPJFJAM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM/action/storage_attestation","attest_author":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM/action/author_attestation","sign_citation":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM/action/citation_signature","submit_replication":"https://pith.science/pith/7BPB43YHCIQE35U5TOLRPJFJAM/action/replication_record"}},"created_at":"2026-07-05T01:53:00.052326+00:00","updated_at":"2026-07-05T01:53:00.052326+00:00"}