{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:G5TAHVFRLEUHWG2TSIKOQSUCOT","short_pith_number":"pith:G5TAHVFR","schema_version":"1.0","canonical_sha256":"376603d4b159287b1b539214e84a8274ebcb5073eb3870f489e0d4d87ddf3936","source":{"kind":"arxiv","id":"2409.02565","version":1},"attestation_state":"computed","paper":{"title":"Efficient Extraction of Noise-Robust Discrete Units from Self-Supervised Speech Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Hugo Van hamme, Jakob Poncelet, Yujun Wang","submitted_at":"2024-09-04T09:31:58Z","abstract_excerpt":"Continuous speech can be converted into a discrete sequence by deriving discrete units from the hidden features of self-supervised learned (SSL) speech models. Although SSL models are becoming larger and trained on more data, they are often sensitive to real-life distortions like additive noise or reverberation, which translates to a shift in discrete units. We propose a parameter-efficient approach to generate noise-robust discrete units from pre-trained SSL models by training a small encoder-decoder model, with or without adapters, to simultaneously denoise and discretise the hidden features"},"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":"2409.02565","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2024-09-04T09:31:58Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"f3da693462886b187352953384f60783b34dee5d7f4bb199a3dc9ce96a18c559","abstract_canon_sha256":"0c9a7addf3bb5b80d40b1cf333e7745174a72b464fb7a1081354a12341c9f411"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:09:47.232830Z","signature_b64":"uQQayh7t0D2Z/3fI2zNqWGlyB5uTwmUjTUDIUjC2JAQkojscSa6q5DHiV7ICWfMRqw8+Bn8DdUBO8mqZjbSJDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"376603d4b159287b1b539214e84a8274ebcb5073eb3870f489e0d4d87ddf3936","last_reissued_at":"2026-07-05T10:09:47.232365Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:09:47.232365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Extraction of Noise-Robust Discrete Units from Self-Supervised Speech Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Hugo Van hamme, Jakob Poncelet, Yujun Wang","submitted_at":"2024-09-04T09:31:58Z","abstract_excerpt":"Continuous speech can be converted into a discrete sequence by deriving discrete units from the hidden features of self-supervised learned (SSL) speech models. Although SSL models are becoming larger and trained on more data, they are often sensitive to real-life distortions like additive noise or reverberation, which translates to a shift in discrete units. We propose a parameter-efficient approach to generate noise-robust discrete units from pre-trained SSL models by training a small encoder-decoder model, with or without adapters, to simultaneously denoise and discretise the hidden features"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.02565","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/2409.02565/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":"2409.02565","created_at":"2026-07-05T10:09:47.232429+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.02565v1","created_at":"2026-07-05T10:09:47.232429+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.02565","created_at":"2026-07-05T10:09:47.232429+00:00"},{"alias_kind":"pith_short_12","alias_value":"G5TAHVFRLEUH","created_at":"2026-07-05T10:09:47.232429+00:00"},{"alias_kind":"pith_short_16","alias_value":"G5TAHVFRLEUHWG2T","created_at":"2026-07-05T10:09:47.232429+00:00"},{"alias_kind":"pith_short_8","alias_value":"G5TAHVFR","created_at":"2026-07-05T10:09:47.232429+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT","json":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT.json","graph_json":"https://pith.science/api/pith-number/G5TAHVFRLEUHWG2TSIKOQSUCOT/graph.json","events_json":"https://pith.science/api/pith-number/G5TAHVFRLEUHWG2TSIKOQSUCOT/events.json","paper":"https://pith.science/paper/G5TAHVFR"},"agent_actions":{"view_html":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT","download_json":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT.json","view_paper":"https://pith.science/paper/G5TAHVFR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.02565&json=true","fetch_graph":"https://pith.science/api/pith-number/G5TAHVFRLEUHWG2TSIKOQSUCOT/graph.json","fetch_events":"https://pith.science/api/pith-number/G5TAHVFRLEUHWG2TSIKOQSUCOT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT/action/storage_attestation","attest_author":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT/action/author_attestation","sign_citation":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT/action/citation_signature","submit_replication":"https://pith.science/pith/G5TAHVFRLEUHWG2TSIKOQSUCOT/action/replication_record"}},"created_at":"2026-07-05T10:09:47.232429+00:00","updated_at":"2026-07-05T10:09:47.232429+00:00"}