{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:62JDGXTLY4SWJGEORX6BDU4UGP","short_pith_number":"pith:62JDGXTL","schema_version":"1.0","canonical_sha256":"f692335e6bc72564988e8dfc11d39433deda659b0672f89886921631bdf2842d","source":{"kind":"arxiv","id":"2110.11499","version":2},"attestation_state":"computed","paper":{"title":"Wav2CLIP: Learning Robust Audio Representations From CLIP","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ho-Hsiang Wu, Juan Pablo Bello, Kundan Kumar, Prem Seetharaman","submitted_at":"2021-10-21T22:00:13Z","abstract_excerpt":"We propose Wav2CLIP, a robust audio representation learning method by distilling from Contrastive Language-Image Pre-training (CLIP). We systematically evaluate Wav2CLIP on a variety of audio tasks including classification, retrieval, and generation, and show that Wav2CLIP can outperform several publicly available pre-trained audio representation algorithms. Wav2CLIP projects audio into a shared embedding space with images and text, which enables multimodal applications such as zero-shot classification, and cross-modal retrieval. Furthermore, Wav2CLIP needs just ~10% of the data to achieve com"},"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":"2110.11499","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2021-10-21T22:00:13Z","cross_cats_sorted":["cs.LG","eess.AS"],"title_canon_sha256":"db9bdb35efddfa5af1369a2eae02894317baaf12f9bdb71e95d13837c1724e11","abstract_canon_sha256":"e2dde1695ac18295b358c4bcd41ac1d5041a6e669992adf73e548df0b960f1ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:56:51.237377Z","signature_b64":"5zbeh5Uoofb094neuCB3qzBB0qpDraN9FMdfi5QP4+Xx+wV7AIR43dwPHUO3HooLfH1QsBWB+ByJctfLTim2Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f692335e6bc72564988e8dfc11d39433deda659b0672f89886921631bdf2842d","last_reissued_at":"2026-07-05T03:56:51.236964Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:56:51.236964Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wav2CLIP: Learning Robust Audio Representations From CLIP","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ho-Hsiang Wu, Juan Pablo Bello, Kundan Kumar, Prem Seetharaman","submitted_at":"2021-10-21T22:00:13Z","abstract_excerpt":"We propose Wav2CLIP, a robust audio representation learning method by distilling from Contrastive Language-Image Pre-training (CLIP). We systematically evaluate Wav2CLIP on a variety of audio tasks including classification, retrieval, and generation, and show that Wav2CLIP can outperform several publicly available pre-trained audio representation algorithms. Wav2CLIP projects audio into a shared embedding space with images and text, which enables multimodal applications such as zero-shot classification, and cross-modal retrieval. Furthermore, Wav2CLIP needs just ~10% of the data to achieve com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.11499","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/2110.11499/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":"2110.11499","created_at":"2026-07-05T03:56:51.237019+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.11499v2","created_at":"2026-07-05T03:56:51.237019+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.11499","created_at":"2026-07-05T03:56:51.237019+00:00"},{"alias_kind":"pith_short_12","alias_value":"62JDGXTLY4SW","created_at":"2026-07-05T03:56:51.237019+00:00"},{"alias_kind":"pith_short_16","alias_value":"62JDGXTLY4SWJGEO","created_at":"2026-07-05T03:56:51.237019+00:00"},{"alias_kind":"pith_short_8","alias_value":"62JDGXTL","created_at":"2026-07-05T03:56:51.237019+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2204.00598","citing_title":"Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language","ref_index":88,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16678","citing_title":"UniCon: Unified Framework for Efficient Contrastive Alignment via Kernels","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP","json":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP.json","graph_json":"https://pith.science/api/pith-number/62JDGXTLY4SWJGEORX6BDU4UGP/graph.json","events_json":"https://pith.science/api/pith-number/62JDGXTLY4SWJGEORX6BDU4UGP/events.json","paper":"https://pith.science/paper/62JDGXTL"},"agent_actions":{"view_html":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP","download_json":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP.json","view_paper":"https://pith.science/paper/62JDGXTL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.11499&json=true","fetch_graph":"https://pith.science/api/pith-number/62JDGXTLY4SWJGEORX6BDU4UGP/graph.json","fetch_events":"https://pith.science/api/pith-number/62JDGXTLY4SWJGEORX6BDU4UGP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP/action/storage_attestation","attest_author":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP/action/author_attestation","sign_citation":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP/action/citation_signature","submit_replication":"https://pith.science/pith/62JDGXTLY4SWJGEORX6BDU4UGP/action/replication_record"}},"created_at":"2026-07-05T03:56:51.237019+00:00","updated_at":"2026-07-05T03:56:51.237019+00:00"}