{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:RVFQAC4AXRGUFSVEC7VOJTXH4K","short_pith_number":"pith:RVFQAC4A","schema_version":"1.0","canonical_sha256":"8d4b000b80bc4d42caa417eae4cee7e2a3e790f2767a949113518621b12e9713","source":{"kind":"arxiv","id":"2002.12764","version":6},"attestation_state":"computed","paper":{"title":"Towards Learning a Universal Non-Semantic Representation of Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Aren Jansen, Dotan Emanuel, Felix de Chaumont Quitry, Ira Shavitt, Joel Shor, Marco Tagliasacchi, Omry Tuval, Oran Lang, Ronnie Maor, Yinnon Haviv","submitted_at":"2020-02-25T21:38:24Z","abstract_excerpt":"The ultimate goal of transfer learning is to reduce labeled data requirements by exploiting a pre-existing embedding model trained for different datasets or tasks. The visual and language communities have established benchmarks to compare embeddings, but the speech community has yet to do so. This paper proposes a benchmark for comparing speech representations on non-semantic tasks, and proposes a representation based on an unsupervised triplet-loss objective. The proposed representation outperforms other representations on the benchmark, and even exceeds state-of-the-art performance on a numb"},"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":"2002.12764","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-02-25T21:38:24Z","cross_cats_sorted":["cs.LG","cs.SD","stat.ML"],"title_canon_sha256":"d7104e194bbb08cdbe078dfa90c75fd9cd4baa47625076c99ad7bc1517808bbf","abstract_canon_sha256":"04abef6f2e9c57e9710e8aa41a194062af88d790485a2092fb2bdf1af8cac3a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:38:28.334152Z","signature_b64":"ztwAsnkySRyYStnZVi1ZHsEirW3oYibbsPgGkerBfvPfWtXxr53nnLIwhZ/hrzI/iRmVbi7H03guZ5CiNoFnDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d4b000b80bc4d42caa417eae4cee7e2a3e790f2767a949113518621b12e9713","last_reissued_at":"2026-07-05T02:38:28.333655Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:38:28.333655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Learning a Universal Non-Semantic Representation of Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Aren Jansen, Dotan Emanuel, Felix de Chaumont Quitry, Ira Shavitt, Joel Shor, Marco Tagliasacchi, Omry Tuval, Oran Lang, Ronnie Maor, Yinnon Haviv","submitted_at":"2020-02-25T21:38:24Z","abstract_excerpt":"The ultimate goal of transfer learning is to reduce labeled data requirements by exploiting a pre-existing embedding model trained for different datasets or tasks. The visual and language communities have established benchmarks to compare embeddings, but the speech community has yet to do so. This paper proposes a benchmark for comparing speech representations on non-semantic tasks, and proposes a representation based on an unsupervised triplet-loss objective. The proposed representation outperforms other representations on the benchmark, and even exceeds state-of-the-art performance on a numb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.12764","kind":"arxiv","version":6},"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/2002.12764/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":"2002.12764","created_at":"2026-07-05T02:38:28.333714+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.12764v6","created_at":"2026-07-05T02:38:28.333714+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.12764","created_at":"2026-07-05T02:38:28.333714+00:00"},{"alias_kind":"pith_short_12","alias_value":"RVFQAC4AXRGU","created_at":"2026-07-05T02:38:28.333714+00:00"},{"alias_kind":"pith_short_16","alias_value":"RVFQAC4AXRGUFSVE","created_at":"2026-07-05T02:38:28.333714+00:00"},{"alias_kind":"pith_short_8","alias_value":"RVFQAC4A","created_at":"2026-07-05T02:38:28.333714+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00387","citing_title":"From Objectives to Applications: Aligning Architectural Biases in Audio Self-Supervised Learning","ref_index":144,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K","json":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K.json","graph_json":"https://pith.science/api/pith-number/RVFQAC4AXRGUFSVEC7VOJTXH4K/graph.json","events_json":"https://pith.science/api/pith-number/RVFQAC4AXRGUFSVEC7VOJTXH4K/events.json","paper":"https://pith.science/paper/RVFQAC4A"},"agent_actions":{"view_html":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K","download_json":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K.json","view_paper":"https://pith.science/paper/RVFQAC4A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.12764&json=true","fetch_graph":"https://pith.science/api/pith-number/RVFQAC4AXRGUFSVEC7VOJTXH4K/graph.json","fetch_events":"https://pith.science/api/pith-number/RVFQAC4AXRGUFSVEC7VOJTXH4K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K/action/storage_attestation","attest_author":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K/action/author_attestation","sign_citation":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K/action/citation_signature","submit_replication":"https://pith.science/pith/RVFQAC4AXRGUFSVEC7VOJTXH4K/action/replication_record"}},"created_at":"2026-07-05T02:38:28.333714+00:00","updated_at":"2026-07-05T02:38:28.333714+00:00"}