{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AYACQVRSQPGLPLOTCPL65BXREZ","short_pith_number":"pith:AYACQVRS","schema_version":"1.0","canonical_sha256":"060028563283ccb7add313d7ee86f12650828b4d36be5ba8a808eae12c31d55d","source":{"kind":"arxiv","id":"2402.01945","version":1},"attestation_state":"computed","paper":{"title":"A Case Study on Filtering for End-to-End Speech Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Antonios Anastasopoulos, Md Mahfuz ibn Alam","submitted_at":"2024-02-02T22:42:33Z","abstract_excerpt":"It is relatively easy to mine a large parallel corpus for any machine learning task, such as speech-to-text or speech-to-speech translation. Although these mined corpora are large in volume, their quality is questionable. This work shows that the simplest filtering technique can trim down these big, noisy datasets to a more manageable, clean dataset. We also show that using this clean dataset can improve the model's performance, as in the case of the multilingual-to-English Speech Translation (ST) model, where, on average, we obtain a 4.65 BLEU score improvement."},"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":"2402.01945","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-02T22:42:33Z","cross_cats_sorted":[],"title_canon_sha256":"d2855773a9c7f02d641de34c7c7be5d699b0197a4c0c8a174909332ba2aa8f5f","abstract_canon_sha256":"20267f0ae4d0f62d28660f406e220196e305e56785f0a53cdb9627d31b13a029"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:41:08.519014Z","signature_b64":"0F5UeyDrsZ4/wX5g4SQuKf0F11LWskKneqRfsbuOfMKqT/0UrBRAkn1SKPqsKZdip+lEYIYSqzx4XT12q9xiBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"060028563283ccb7add313d7ee86f12650828b4d36be5ba8a808eae12c31d55d","last_reissued_at":"2026-07-05T07:41:08.518582Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:41:08.518582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Case Study on Filtering for End-to-End Speech Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Antonios Anastasopoulos, Md Mahfuz ibn Alam","submitted_at":"2024-02-02T22:42:33Z","abstract_excerpt":"It is relatively easy to mine a large parallel corpus for any machine learning task, such as speech-to-text or speech-to-speech translation. Although these mined corpora are large in volume, their quality is questionable. This work shows that the simplest filtering technique can trim down these big, noisy datasets to a more manageable, clean dataset. We also show that using this clean dataset can improve the model's performance, as in the case of the multilingual-to-English Speech Translation (ST) model, where, on average, we obtain a 4.65 BLEU score improvement."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01945","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/2402.01945/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":"2402.01945","created_at":"2026-07-05T07:41:08.518647+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.01945v1","created_at":"2026-07-05T07:41:08.518647+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01945","created_at":"2026-07-05T07:41:08.518647+00:00"},{"alias_kind":"pith_short_12","alias_value":"AYACQVRSQPGL","created_at":"2026-07-05T07:41:08.518647+00:00"},{"alias_kind":"pith_short_16","alias_value":"AYACQVRSQPGLPLOT","created_at":"2026-07-05T07:41:08.518647+00:00"},{"alias_kind":"pith_short_8","alias_value":"AYACQVRS","created_at":"2026-07-05T07:41:08.518647+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13507","citing_title":"Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ","json":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ.json","graph_json":"https://pith.science/api/pith-number/AYACQVRSQPGLPLOTCPL65BXREZ/graph.json","events_json":"https://pith.science/api/pith-number/AYACQVRSQPGLPLOTCPL65BXREZ/events.json","paper":"https://pith.science/paper/AYACQVRS"},"agent_actions":{"view_html":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ","download_json":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ.json","view_paper":"https://pith.science/paper/AYACQVRS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.01945&json=true","fetch_graph":"https://pith.science/api/pith-number/AYACQVRSQPGLPLOTCPL65BXREZ/graph.json","fetch_events":"https://pith.science/api/pith-number/AYACQVRSQPGLPLOTCPL65BXREZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ/action/storage_attestation","attest_author":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ/action/author_attestation","sign_citation":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ/action/citation_signature","submit_replication":"https://pith.science/pith/AYACQVRSQPGLPLOTCPL65BXREZ/action/replication_record"}},"created_at":"2026-07-05T07:41:08.518647+00:00","updated_at":"2026-07-05T07:41:08.518647+00:00"}