{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LHWUAMQSPDQJB2RLH6O3QZRX7D","short_pith_number":"pith:LHWUAMQS","schema_version":"1.0","canonical_sha256":"59ed40321278e090ea2b3f9db86637f8c3db95dc6932049cdfdb0a2639e8043c","source":{"kind":"arxiv","id":"2505.23077","version":1},"attestation_state":"computed","paper":{"title":"Contextualized Automatic Speech Recognition with Dynamic Vocabulary Prediction and Activation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Kaixun Huang, Lei Xie, Linju Yang, Wei Ren, Zhennan Lin","submitted_at":"2025-05-29T04:31:33Z","abstract_excerpt":"Deep biasing improves automatic speech recognition (ASR) performance by incorporating contextual phrases. However, most existing methods enhance subwords in a contextual phrase as independent units, potentially compromising contextual phrase integrity, leading to accuracy reduction. In this paper, we propose an encoder-based phrase-level contextualized ASR method that leverages dynamic vocabulary prediction and activation. We introduce architectural optimizations and integrate a bias loss to extend phrase-level predictions based on frame-level outputs. We also introduce a confidence-activated "},"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":"2505.23077","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-05-29T04:31:33Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"eb705ceaac3e6207cb0b0000676f5893bac82b4f824c1e727e37a2ade74fd73e","abstract_canon_sha256":"8cb544c2eb35a971b220dbd2639b7b5c073fdfabf749b470b1a11100897d2343"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:11.949293Z","signature_b64":"H14628iwpczBVH1AWcUJLYIhZQHcFrkz+LS8CuyO0a1w1ztaJExMxbZo/quJksfh2Vw0q3fmfN18+PsqT7dsDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59ed40321278e090ea2b3f9db86637f8c3db95dc6932049cdfdb0a2639e8043c","last_reissued_at":"2026-07-05T11:12:11.948832Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:11.948832Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contextualized Automatic Speech Recognition with Dynamic Vocabulary Prediction and Activation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Kaixun Huang, Lei Xie, Linju Yang, Wei Ren, Zhennan Lin","submitted_at":"2025-05-29T04:31:33Z","abstract_excerpt":"Deep biasing improves automatic speech recognition (ASR) performance by incorporating contextual phrases. However, most existing methods enhance subwords in a contextual phrase as independent units, potentially compromising contextual phrase integrity, leading to accuracy reduction. In this paper, we propose an encoder-based phrase-level contextualized ASR method that leverages dynamic vocabulary prediction and activation. We introduce architectural optimizations and integrate a bias loss to extend phrase-level predictions based on frame-level outputs. We also introduce a confidence-activated "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23077","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/2505.23077/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":"2505.23077","created_at":"2026-07-05T11:12:11.948884+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.23077v1","created_at":"2026-07-05T11:12:11.948884+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23077","created_at":"2026-07-05T11:12:11.948884+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHWUAMQSPDQJ","created_at":"2026-07-05T11:12:11.948884+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHWUAMQSPDQJB2RL","created_at":"2026-07-05T11:12:11.948884+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHWUAMQS","created_at":"2026-07-05T11:12:11.948884+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.23077","citing_title":"Contextualized Automatic Speech Recognition with Dynamic Vocabulary Prediction and Activation","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D","json":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D.json","graph_json":"https://pith.science/api/pith-number/LHWUAMQSPDQJB2RLH6O3QZRX7D/graph.json","events_json":"https://pith.science/api/pith-number/LHWUAMQSPDQJB2RLH6O3QZRX7D/events.json","paper":"https://pith.science/paper/LHWUAMQS"},"agent_actions":{"view_html":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D","download_json":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D.json","view_paper":"https://pith.science/paper/LHWUAMQS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.23077&json=true","fetch_graph":"https://pith.science/api/pith-number/LHWUAMQSPDQJB2RLH6O3QZRX7D/graph.json","fetch_events":"https://pith.science/api/pith-number/LHWUAMQSPDQJB2RLH6O3QZRX7D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D/action/storage_attestation","attest_author":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D/action/author_attestation","sign_citation":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D/action/citation_signature","submit_replication":"https://pith.science/pith/LHWUAMQSPDQJB2RLH6O3QZRX7D/action/replication_record"}},"created_at":"2026-07-05T11:12:11.948884+00:00","updated_at":"2026-07-05T11:12:11.948884+00:00"}