{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UJCJXOF6EN2HJAMDUOXNTH7L2R","short_pith_number":"pith:UJCJXOF6","schema_version":"1.0","canonical_sha256":"a2449bb8be2374748183a3aed99febd472080f69ce2cefb7f685dc58781db7a6","source":{"kind":"arxiv","id":"2506.07237","version":2},"attestation_state":"computed","paper":{"title":"Multi-Distillation from Speech and Music Representation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Fabian Ritter-Gutierrez, Hung-yi Lee, Jui-Chiang Wei, Yi-Cheng Lin","submitted_at":"2025-06-08T17:45:46Z","abstract_excerpt":"Real-world audio often mixes speech and music, yet models typically handle only one domain. This paper introduces a multi-teacher distillation framework that unifies speech and music models into a single one while significantly reducing model size. Our approach leverages the strengths of domain-specific teacher models, such as HuBERT for speech and MERT for music, and explores various strategies to balance both domains. Experiments across diverse tasks demonstrate that our model matches the performance of domain-specific models, showing the effectiveness of cross-domain distillation. Additiona"},"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":"2506.07237","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-06-08T17:45:46Z","cross_cats_sorted":[],"title_canon_sha256":"ffca6ecd30ab3ab3fad230f9c6b95c9066a64f5b7af44a99bfb1e9facda58e49","abstract_canon_sha256":"bfd3fe3f60b4b6a753c71e5e816e8e31d9d15ad121fa2b10cfff1e7f6ea7d5d7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:53.555943Z","signature_b64":"lMcgNUjU2fJ6ubhmIrdy4qMFtuF3wBnZPLHY7PXukVymNAvVwSr8V1N297PX13fy/XbBBqsqnPnrnB/fIiO/Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2449bb8be2374748183a3aed99febd472080f69ce2cefb7f685dc58781db7a6","last_reissued_at":"2026-07-05T11:19:53.555491Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:53.555491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Distillation from Speech and Music Representation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Fabian Ritter-Gutierrez, Hung-yi Lee, Jui-Chiang Wei, Yi-Cheng Lin","submitted_at":"2025-06-08T17:45:46Z","abstract_excerpt":"Real-world audio often mixes speech and music, yet models typically handle only one domain. This paper introduces a multi-teacher distillation framework that unifies speech and music models into a single one while significantly reducing model size. Our approach leverages the strengths of domain-specific teacher models, such as HuBERT for speech and MERT for music, and explores various strategies to balance both domains. Experiments across diverse tasks demonstrate that our model matches the performance of domain-specific models, showing the effectiveness of cross-domain distillation. Additiona"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07237","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/2506.07237/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":"2506.07237","created_at":"2026-07-05T11:19:53.555547+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.07237v2","created_at":"2026-07-05T11:19:53.555547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07237","created_at":"2026-07-05T11:19:53.555547+00:00"},{"alias_kind":"pith_short_12","alias_value":"UJCJXOF6EN2H","created_at":"2026-07-05T11:19:53.555547+00:00"},{"alias_kind":"pith_short_16","alias_value":"UJCJXOF6EN2HJAMD","created_at":"2026-07-05T11:19:53.555547+00:00"},{"alias_kind":"pith_short_8","alias_value":"UJCJXOF6","created_at":"2026-07-05T11:19:53.555547+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.01732","citing_title":"EGAD: Entropy-Guided Adaptive Distillation for Token-Level Knowledge Transfer","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R","json":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R.json","graph_json":"https://pith.science/api/pith-number/UJCJXOF6EN2HJAMDUOXNTH7L2R/graph.json","events_json":"https://pith.science/api/pith-number/UJCJXOF6EN2HJAMDUOXNTH7L2R/events.json","paper":"https://pith.science/paper/UJCJXOF6"},"agent_actions":{"view_html":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R","download_json":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R.json","view_paper":"https://pith.science/paper/UJCJXOF6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.07237&json=true","fetch_graph":"https://pith.science/api/pith-number/UJCJXOF6EN2HJAMDUOXNTH7L2R/graph.json","fetch_events":"https://pith.science/api/pith-number/UJCJXOF6EN2HJAMDUOXNTH7L2R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R/action/storage_attestation","attest_author":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R/action/author_attestation","sign_citation":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R/action/citation_signature","submit_replication":"https://pith.science/pith/UJCJXOF6EN2HJAMDUOXNTH7L2R/action/replication_record"}},"created_at":"2026-07-05T11:19:53.555547+00:00","updated_at":"2026-07-05T11:19:53.555547+00:00"}