{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HSNA5MCY7MFWXEKWF7VXU5PSUR","short_pith_number":"pith:HSNA5MCY","schema_version":"1.0","canonical_sha256":"3c9a0eb058fb0b6b91562feb7a75f2a4618edb186058c7a9e8d4a9a9a939d47d","source":{"kind":"arxiv","id":"2506.02181","version":1},"attestation_state":"computed","paper":{"title":"Echoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attribution","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dennis Fucci, Luisa Bentivogli, Marco Gaido, Matteo Negri, Mauro Cettolo","submitted_at":"2025-06-02T19:11:16Z","abstract_excerpt":"Despite significant advances in ASR, the specific acoustic cues models rely on remain unclear. Prior studies have examined such cues on a limited set of phonemes and outdated models. In this work, we apply a feature attribution technique to identify the relevant acoustic cues for a modern Conformer-based ASR system. By analyzing plosives, fricatives, and vowels, we assess how feature attributions align with their acoustic properties in the time and frequency domains, also essential for human speech perception. Our findings show that the ASR model relies on vowels' full time spans, particularly"},"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":"2506.02181","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T19:11:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"76d5ae4bcd0913bd4d4fb69f619893b731d13b6ce5b30c05c9a7c2c526d07e8d","abstract_canon_sha256":"98c0b475286063097fd36db641b4b618662db269a0f8d09a5eaa58be9fe350e0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:33.855996Z","signature_b64":"vyTRdYC6jGTmvsqU9hIGXgTym8uXely/MsoEbZMk79m+hGPcNzAAtWRRlUlDl3I5Lsvwr16eGAkVBokd+0PpBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c9a0eb058fb0b6b91562feb7a75f2a4618edb186058c7a9e8d4a9a9a939d47d","last_reissued_at":"2026-07-05T11:14:33.855450Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:33.855450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Echoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attribution","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dennis Fucci, Luisa Bentivogli, Marco Gaido, Matteo Negri, Mauro Cettolo","submitted_at":"2025-06-02T19:11:16Z","abstract_excerpt":"Despite significant advances in ASR, the specific acoustic cues models rely on remain unclear. Prior studies have examined such cues on a limited set of phonemes and outdated models. In this work, we apply a feature attribution technique to identify the relevant acoustic cues for a modern Conformer-based ASR system. By analyzing plosives, fricatives, and vowels, we assess how feature attributions align with their acoustic properties in the time and frequency domains, also essential for human speech perception. Our findings show that the ASR model relies on vowels' full time spans, particularly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02181","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/2506.02181/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":"2506.02181","created_at":"2026-07-05T11:14:33.855524+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.02181v1","created_at":"2026-07-05T11:14:33.855524+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02181","created_at":"2026-07-05T11:14:33.855524+00:00"},{"alias_kind":"pith_short_12","alias_value":"HSNA5MCY7MFW","created_at":"2026-07-05T11:14:33.855524+00:00"},{"alias_kind":"pith_short_16","alias_value":"HSNA5MCY7MFWXEKW","created_at":"2026-07-05T11:14:33.855524+00:00"},{"alias_kind":"pith_short_8","alias_value":"HSNA5MCY","created_at":"2026-07-05T11:14:33.855524+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02181","citing_title":"Echoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attribution","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR","json":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR.json","graph_json":"https://pith.science/api/pith-number/HSNA5MCY7MFWXEKWF7VXU5PSUR/graph.json","events_json":"https://pith.science/api/pith-number/HSNA5MCY7MFWXEKWF7VXU5PSUR/events.json","paper":"https://pith.science/paper/HSNA5MCY"},"agent_actions":{"view_html":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR","download_json":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR.json","view_paper":"https://pith.science/paper/HSNA5MCY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.02181&json=true","fetch_graph":"https://pith.science/api/pith-number/HSNA5MCY7MFWXEKWF7VXU5PSUR/graph.json","fetch_events":"https://pith.science/api/pith-number/HSNA5MCY7MFWXEKWF7VXU5PSUR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR/action/storage_attestation","attest_author":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR/action/author_attestation","sign_citation":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR/action/citation_signature","submit_replication":"https://pith.science/pith/HSNA5MCY7MFWXEKWF7VXU5PSUR/action/replication_record"}},"created_at":"2026-07-05T11:14:33.855524+00:00","updated_at":"2026-07-05T11:14:33.855524+00:00"}