{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GWZZHCJSX5GIJKFKUU45YKFPGZ","short_pith_number":"pith:GWZZHCJS","schema_version":"1.0","canonical_sha256":"35b3938932bf4c84a8aaa539dc28af3662a37b6bc7572d4f7a53aa854b3ee236","source":{"kind":"arxiv","id":"2505.22005","version":1},"attestation_state":"computed","paper":{"title":"Leveraging LLM for Stuttering Speech: A Unified Architecture Bridging Recognition and Event Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Jing Deng, Jintao Kang, Rong Zheng, Shangkun Huang","submitted_at":"2025-05-28T06:12:19Z","abstract_excerpt":"The performance bottleneck of Automatic Speech Recognition (ASR) in stuttering speech scenarios has limited its applicability in domains such as speech rehabilitation. This paper proposed an LLM-driven ASR-SED multi-task learning framework that jointly optimized the ASR and Stuttering Event Detection (SED) tasks. We proposed a dynamic interaction mechanism where the ASR branch leveraged CTC-generated soft prompts to assist LLM context modeling, while the SED branch output stutter embeddings to enhance LLM comprehension of stuttered speech. We incorporated contrastive learning to strengthen the"},"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":"2505.22005","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-05-28T06:12:19Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"70618963be0277541daab87d6d45870ff9e2fc58a0a6bb63c154046fbf1ae171","abstract_canon_sha256":"1d8404011954f4a3ce012b72a02592b8057a0afa7f04085b159842b14042f8ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:07.807822Z","signature_b64":"cJoJgwhTRg+boWGBPPeEv9QwkS+hdbxIf/7ZiIypIh+NJDFwHCngrKwYwNPS502SguS2VU6Lys1mJ5K843vFDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35b3938932bf4c84a8aaa539dc28af3662a37b6bc7572d4f7a53aa854b3ee236","last_reissued_at":"2026-07-05T11:11:07.807135Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:07.807135Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging LLM for Stuttering Speech: A Unified Architecture Bridging Recognition and Event Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Jing Deng, Jintao Kang, Rong Zheng, Shangkun Huang","submitted_at":"2025-05-28T06:12:19Z","abstract_excerpt":"The performance bottleneck of Automatic Speech Recognition (ASR) in stuttering speech scenarios has limited its applicability in domains such as speech rehabilitation. This paper proposed an LLM-driven ASR-SED multi-task learning framework that jointly optimized the ASR and Stuttering Event Detection (SED) tasks. We proposed a dynamic interaction mechanism where the ASR branch leveraged CTC-generated soft prompts to assist LLM context modeling, while the SED branch output stutter embeddings to enhance LLM comprehension of stuttered speech. We incorporated contrastive learning to strengthen the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22005","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/2505.22005/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":"2505.22005","created_at":"2026-07-05T11:11:07.807215+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.22005v1","created_at":"2026-07-05T11:11:07.807215+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22005","created_at":"2026-07-05T11:11:07.807215+00:00"},{"alias_kind":"pith_short_12","alias_value":"GWZZHCJSX5GI","created_at":"2026-07-05T11:11:07.807215+00:00"},{"alias_kind":"pith_short_16","alias_value":"GWZZHCJSX5GIJKFK","created_at":"2026-07-05T11:11:07.807215+00:00"},{"alias_kind":"pith_short_8","alias_value":"GWZZHCJS","created_at":"2026-07-05T11:11:07.807215+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22005","citing_title":"Leveraging LLM for Stuttering Speech: A Unified Architecture Bridging Recognition and Event Detection","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ","json":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ.json","graph_json":"https://pith.science/api/pith-number/GWZZHCJSX5GIJKFKUU45YKFPGZ/graph.json","events_json":"https://pith.science/api/pith-number/GWZZHCJSX5GIJKFKUU45YKFPGZ/events.json","paper":"https://pith.science/paper/GWZZHCJS"},"agent_actions":{"view_html":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ","download_json":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ.json","view_paper":"https://pith.science/paper/GWZZHCJS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.22005&json=true","fetch_graph":"https://pith.science/api/pith-number/GWZZHCJSX5GIJKFKUU45YKFPGZ/graph.json","fetch_events":"https://pith.science/api/pith-number/GWZZHCJSX5GIJKFKUU45YKFPGZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ/action/storage_attestation","attest_author":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ/action/author_attestation","sign_citation":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ/action/citation_signature","submit_replication":"https://pith.science/pith/GWZZHCJSX5GIJKFKUU45YKFPGZ/action/replication_record"}},"created_at":"2026-07-05T11:11:07.807215+00:00","updated_at":"2026-07-05T11:11:07.807215+00:00"}