{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Q3F2ZTCF6JSMAKNQNEMZAXT336","short_pith_number":"pith:Q3F2ZTCF","schema_version":"1.0","canonical_sha256":"86cbaccc45f264c029b06919905e7bdf9637b9e963971ffa928b11ec11e50285","source":{"kind":"arxiv","id":"2306.07944","version":1},"attestation_state":"computed","paper":{"title":"Speech-to-Text Adapter and Speech-to-Entity Retriever Augmented LLMs for Speech Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"eess.AS","authors_text":"Dian Yu, Hagen Soltau, Izhak Shafran, Laurent El Shafey, Mingqiu Wang, Wei Han, Yuan Cao","submitted_at":"2023-06-08T22:33:22Z","abstract_excerpt":"Large Language Models (LLMs) have been applied in the speech domain, often incurring a performance drop due to misaligned between speech and language representations. To bridge this gap, we propose a joint speech and language model (SLM) using a Speech2Text adapter, which maps speech into text token embedding space without speech information loss. Additionally, using a CTC-based blank-filtering, we can reduce the speech sequence length to that of text. In speech MultiWoz dataset (DSTC11 challenge), SLM largely improves the dialog state tracking (DST) performance (24.7% to 28.4% accuracy). Furt"},"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":"2306.07944","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2023-06-08T22:33:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"7628fcacce49793a085cf26b543c1abb3ed74dd6d5e6944fc558253095d352b2","abstract_canon_sha256":"7bb6e9a4baa1b5d23dff94700292d3d541db2ae9df5d7cad62a40406df21b608"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:20:11.259384Z","signature_b64":"HjTdGYw0+COp530DKcolra+BEV4hpogSa6RcNw0JEkPUW+SLFApD8nU9hurvtm2fpQ8raIcaJ+VG8Yvg3juDAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86cbaccc45f264c029b06919905e7bdf9637b9e963971ffa928b11ec11e50285","last_reissued_at":"2026-07-05T06:20:11.258915Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:20:11.258915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Speech-to-Text Adapter and Speech-to-Entity Retriever Augmented LLMs for Speech Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"eess.AS","authors_text":"Dian Yu, Hagen Soltau, Izhak Shafran, Laurent El Shafey, Mingqiu Wang, Wei Han, Yuan Cao","submitted_at":"2023-06-08T22:33:22Z","abstract_excerpt":"Large Language Models (LLMs) have been applied in the speech domain, often incurring a performance drop due to misaligned between speech and language representations. To bridge this gap, we propose a joint speech and language model (SLM) using a Speech2Text adapter, which maps speech into text token embedding space without speech information loss. Additionally, using a CTC-based blank-filtering, we can reduce the speech sequence length to that of text. In speech MultiWoz dataset (DSTC11 challenge), SLM largely improves the dialog state tracking (DST) performance (24.7% to 28.4% accuracy). Furt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07944","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/2306.07944/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":"2306.07944","created_at":"2026-07-05T06:20:11.258982+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07944v1","created_at":"2026-07-05T06:20:11.258982+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07944","created_at":"2026-07-05T06:20:11.258982+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q3F2ZTCF6JSM","created_at":"2026-07-05T06:20:11.258982+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q3F2ZTCF6JSMAKNQ","created_at":"2026-07-05T06:20:11.258982+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q3F2ZTCF","created_at":"2026-07-05T06:20:11.258982+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/Q3F2ZTCF6JSMAKNQNEMZAXT336","json":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336.json","graph_json":"https://pith.science/api/pith-number/Q3F2ZTCF6JSMAKNQNEMZAXT336/graph.json","events_json":"https://pith.science/api/pith-number/Q3F2ZTCF6JSMAKNQNEMZAXT336/events.json","paper":"https://pith.science/paper/Q3F2ZTCF"},"agent_actions":{"view_html":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336","download_json":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336.json","view_paper":"https://pith.science/paper/Q3F2ZTCF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07944&json=true","fetch_graph":"https://pith.science/api/pith-number/Q3F2ZTCF6JSMAKNQNEMZAXT336/graph.json","fetch_events":"https://pith.science/api/pith-number/Q3F2ZTCF6JSMAKNQNEMZAXT336/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336/action/storage_attestation","attest_author":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336/action/author_attestation","sign_citation":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336/action/citation_signature","submit_replication":"https://pith.science/pith/Q3F2ZTCF6JSMAKNQNEMZAXT336/action/replication_record"}},"created_at":"2026-07-05T06:20:11.258982+00:00","updated_at":"2026-07-05T06:20:11.258982+00:00"}