{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QVQDJOKUH7JAMZY3TRVRNZGQS4","short_pith_number":"pith:QVQDJOKU","schema_version":"1.0","canonical_sha256":"856034b9543fd206671b9c6b16e4d0973dce6f4e931b7b6c51bacd9343fe4148","source":{"kind":"arxiv","id":"2309.12242","version":1},"attestation_state":"computed","paper":{"title":"Weakly-supervised Automated Audio Captioning via text only training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Theodoros Kouzelis, Vassilis Katsouros","submitted_at":"2023-09-21T16:40:46Z","abstract_excerpt":"In recent years, datasets of paired audio and captions have enabled remarkable success in automatically generating descriptions for audio clips, namely Automated Audio Captioning (AAC). However, it is labor-intensive and time-consuming to collect a sufficient number of paired audio and captions. Motivated by the recent advances in Contrastive Language-Audio Pretraining (CLAP), we propose a weakly-supervised approach to train an AAC model assuming only text data and a pre-trained CLAP model, alleviating the need for paired target data. Our approach leverages the similarity between audio and tex"},"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":"2309.12242","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2023-09-21T16:40:46Z","cross_cats_sorted":["cs.LG","eess.AS"],"title_canon_sha256":"7dfc4223169f77258dae24adc457c962b53125581bd0742348f6e99c7e463521","abstract_canon_sha256":"4837004a25073b77a44b667ad8a7e09c9b39df52a9055c67c855fe9cbffb2d95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:53:00.670796Z","signature_b64":"PXU9B4uGTye5Uw7kJ5zYVtZju2Tna9oFdIc4BfD/UVvLJzSIeXMRrRyIYfq+1qvycPzlkSjpmh0bTF+nimYYAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"856034b9543fd206671b9c6b16e4d0973dce6f4e931b7b6c51bacd9343fe4148","last_reissued_at":"2026-07-05T06:53:00.670282Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:53:00.670282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Weakly-supervised Automated Audio Captioning via text only training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Theodoros Kouzelis, Vassilis Katsouros","submitted_at":"2023-09-21T16:40:46Z","abstract_excerpt":"In recent years, datasets of paired audio and captions have enabled remarkable success in automatically generating descriptions for audio clips, namely Automated Audio Captioning (AAC). However, it is labor-intensive and time-consuming to collect a sufficient number of paired audio and captions. Motivated by the recent advances in Contrastive Language-Audio Pretraining (CLAP), we propose a weakly-supervised approach to train an AAC model assuming only text data and a pre-trained CLAP model, alleviating the need for paired target data. Our approach leverages the similarity between audio and tex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.12242","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/2309.12242/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":"2309.12242","created_at":"2026-07-05T06:53:00.670349+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.12242v1","created_at":"2026-07-05T06:53:00.670349+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.12242","created_at":"2026-07-05T06:53:00.670349+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVQDJOKUH7JA","created_at":"2026-07-05T06:53:00.670349+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVQDJOKUH7JAMZY3","created_at":"2026-07-05T06:53:00.670349+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVQDJOKU","created_at":"2026-07-05T06:53:00.670349+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.04376","citing_title":"Domain Adaptation Method and Modality Gap Impact in Audio-Text Models for Prototypical Sound Classification","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4","json":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4.json","graph_json":"https://pith.science/api/pith-number/QVQDJOKUH7JAMZY3TRVRNZGQS4/graph.json","events_json":"https://pith.science/api/pith-number/QVQDJOKUH7JAMZY3TRVRNZGQS4/events.json","paper":"https://pith.science/paper/QVQDJOKU"},"agent_actions":{"view_html":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4","download_json":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4.json","view_paper":"https://pith.science/paper/QVQDJOKU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.12242&json=true","fetch_graph":"https://pith.science/api/pith-number/QVQDJOKUH7JAMZY3TRVRNZGQS4/graph.json","fetch_events":"https://pith.science/api/pith-number/QVQDJOKUH7JAMZY3TRVRNZGQS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4/action/storage_attestation","attest_author":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4/action/author_attestation","sign_citation":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4/action/citation_signature","submit_replication":"https://pith.science/pith/QVQDJOKUH7JAMZY3TRVRNZGQS4/action/replication_record"}},"created_at":"2026-07-05T06:53:00.670349+00:00","updated_at":"2026-07-05T06:53:00.670349+00:00"}