{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:423333Z7UBWF6JUAUIUEQQ5I3X","short_pith_number":"pith:423333Z7","schema_version":"1.0","canonical_sha256":"e6b7bdef3fa06c5f2680a2284843a8ddefb5f41ad4488301dfe0685f1f5b57e6","source":{"kind":"arxiv","id":"2510.25955","version":4},"attestation_state":"computed","paper":{"title":"SPEAR: A Unified SSL Framework for Learning Speech and Audio Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Baoxiang Li, Chao Zhang, Phil Woodland, Wen Wu, Xiaoyu Yang, Yifan Yang, Zengrui Jin, Ziyun Cui","submitted_at":"2025-10-29T20:53:12Z","abstract_excerpt":"Self-supervised learning (SSL) has significantly advanced acoustic representation learning. However, most existing models are optimised for either speech or audio event understanding, resulting in a persistent gap between these two domains. We address this gap with SPEAR (SPEech and Audio Representations), a self-supervised framework that distils complementary knowledge from a speech-focused SSL teacher and a general-audio SSL teacher into a single unified model. SPEAR applies multi-codebook vector quantisation to continuous teacher representations to produce fine-grained discrete tokens that "},"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":"2510.25955","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-10-29T20:53:12Z","cross_cats_sorted":[],"title_canon_sha256":"b03aaa73c5ad62736d7e57fc4cc7b31ca534c1020fa3ce36a80b177bac146719","abstract_canon_sha256":"36c71ec80ea3b00f6387f720f1964c6da8dc090f534c4765528fee9bd77d07a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:13:49.105512Z","signature_b64":"IYSq5aWUMhuTWu9vbrFNJWTylKed6q1+zO13iDWOmetNNAiA+CIsvARW0YlZgMPGONGpGaDgWjpfDRBAZ7IPCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e6b7bdef3fa06c5f2680a2284843a8ddefb5f41ad4488301dfe0685f1f5b57e6","last_reissued_at":"2026-06-23T03:13:49.104984Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:13:49.104984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPEAR: A Unified SSL Framework for Learning Speech and Audio Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Baoxiang Li, Chao Zhang, Phil Woodland, Wen Wu, Xiaoyu Yang, Yifan Yang, Zengrui Jin, Ziyun Cui","submitted_at":"2025-10-29T20:53:12Z","abstract_excerpt":"Self-supervised learning (SSL) has significantly advanced acoustic representation learning. However, most existing models are optimised for either speech or audio event understanding, resulting in a persistent gap between these two domains. We address this gap with SPEAR (SPEech and Audio Representations), a self-supervised framework that distils complementary knowledge from a speech-focused SSL teacher and a general-audio SSL teacher into a single unified model. SPEAR applies multi-codebook vector quantisation to continuous teacher representations to produce fine-grained discrete tokens that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.25955","kind":"arxiv","version":4},"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/2510.25955/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":"2510.25955","created_at":"2026-06-23T03:13:49.105060+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.25955v4","created_at":"2026-06-23T03:13:49.105060+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.25955","created_at":"2026-06-23T03:13:49.105060+00:00"},{"alias_kind":"pith_short_12","alias_value":"423333Z7UBWF","created_at":"2026-06-23T03:13:49.105060+00:00"},{"alias_kind":"pith_short_16","alias_value":"423333Z7UBWF6JUA","created_at":"2026-06-23T03:13:49.105060+00:00"},{"alias_kind":"pith_short_8","alias_value":"423333Z7","created_at":"2026-06-23T03:13:49.105060+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.00387","citing_title":"From Objectives to Applications: Aligning Architectural Biases in Audio Self-Supervised Learning","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00251","citing_title":"Alethia: A Foundational Encoder for Voice Deepfakes","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X","json":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X.json","graph_json":"https://pith.science/api/pith-number/423333Z7UBWF6JUAUIUEQQ5I3X/graph.json","events_json":"https://pith.science/api/pith-number/423333Z7UBWF6JUAUIUEQQ5I3X/events.json","paper":"https://pith.science/paper/423333Z7"},"agent_actions":{"view_html":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X","download_json":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X.json","view_paper":"https://pith.science/paper/423333Z7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.25955&json=true","fetch_graph":"https://pith.science/api/pith-number/423333Z7UBWF6JUAUIUEQQ5I3X/graph.json","fetch_events":"https://pith.science/api/pith-number/423333Z7UBWF6JUAUIUEQQ5I3X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X/action/storage_attestation","attest_author":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X/action/author_attestation","sign_citation":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X/action/citation_signature","submit_replication":"https://pith.science/pith/423333Z7UBWF6JUAUIUEQQ5I3X/action/replication_record"}},"created_at":"2026-06-23T03:13:49.105060+00:00","updated_at":"2026-06-23T03:13:49.105060+00:00"}