{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JSNG2FOVFG7UKGYF3PV5XNTCUL","short_pith_number":"pith:JSNG2FOV","schema_version":"1.0","canonical_sha256":"4c9a6d15d529bf451b05dbebdbb662a2c01a1463cd41d91080db1a3d7d647086","source":{"kind":"arxiv","id":"2102.03207","version":3},"attestation_state":"computed","paper":{"title":"Real-time Denoising and Dereverberation with Tiny Recurrent U-Net","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Dongsuk Jeon, Hoon Heo, Hyeong-Seok Choi, Jie Hwan Lee, Kyogu Lee, Sungjin Park","submitted_at":"2021-02-05T14:46:41Z","abstract_excerpt":"Modern deep learning-based models have seen outstanding performance improvement with speech enhancement tasks. The number of parameters of state-of-the-art models, however, is often too large to be deployed on devices for real-world applications. To this end, we propose Tiny Recurrent U-Net (TRU-Net), a lightweight online inference model that matches the performance of current state-of-the-art models. The size of the quantized version of TRU-Net is 362 kilobytes, which is small enough to be deployed on edge devices. In addition, we combine the small-sized model with a new masking method called"},"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":"2102.03207","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2021-02-05T14:46:41Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"885c0b2f83c7e55c81f5b3768544a9538f64a6f3ee8a1adea3f2f40f30758eb4","abstract_canon_sha256":"cabec1d2fde0bf0d8ea84ed9d9a7396cc4a3f1e487347d0d63ee155aa9c95073"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:51:38.851344Z","signature_b64":"Uph4e9Bow9jrgQTyyJY2l6XIA1L0C3zcz+S2zfMsglOAhbHknJu4hvr/uJjRx7lNJn3LjLwDFrwYfGnEn2v1CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c9a6d15d529bf451b05dbebdbb662a2c01a1463cd41d91080db1a3d7d647086","last_reissued_at":"2026-07-05T02:51:38.850896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:51:38.850896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Real-time Denoising and Dereverberation with Tiny Recurrent U-Net","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Dongsuk Jeon, Hoon Heo, Hyeong-Seok Choi, Jie Hwan Lee, Kyogu Lee, Sungjin Park","submitted_at":"2021-02-05T14:46:41Z","abstract_excerpt":"Modern deep learning-based models have seen outstanding performance improvement with speech enhancement tasks. The number of parameters of state-of-the-art models, however, is often too large to be deployed on devices for real-world applications. To this end, we propose Tiny Recurrent U-Net (TRU-Net), a lightweight online inference model that matches the performance of current state-of-the-art models. The size of the quantized version of TRU-Net is 362 kilobytes, which is small enough to be deployed on edge devices. In addition, we combine the small-sized model with a new masking method called"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.03207","kind":"arxiv","version":3},"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/2102.03207/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":"2102.03207","created_at":"2026-07-05T02:51:38.850960+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.03207v3","created_at":"2026-07-05T02:51:38.850960+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.03207","created_at":"2026-07-05T02:51:38.850960+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSNG2FOVFG7U","created_at":"2026-07-05T02:51:38.850960+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSNG2FOVFG7UKGYF","created_at":"2026-07-05T02:51:38.850960+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSNG2FOV","created_at":"2026-07-05T02:51:38.850960+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/JSNG2FOVFG7UKGYF3PV5XNTCUL","json":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL.json","graph_json":"https://pith.science/api/pith-number/JSNG2FOVFG7UKGYF3PV5XNTCUL/graph.json","events_json":"https://pith.science/api/pith-number/JSNG2FOVFG7UKGYF3PV5XNTCUL/events.json","paper":"https://pith.science/paper/JSNG2FOV"},"agent_actions":{"view_html":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL","download_json":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL.json","view_paper":"https://pith.science/paper/JSNG2FOV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.03207&json=true","fetch_graph":"https://pith.science/api/pith-number/JSNG2FOVFG7UKGYF3PV5XNTCUL/graph.json","fetch_events":"https://pith.science/api/pith-number/JSNG2FOVFG7UKGYF3PV5XNTCUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL/action/storage_attestation","attest_author":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL/action/author_attestation","sign_citation":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL/action/citation_signature","submit_replication":"https://pith.science/pith/JSNG2FOVFG7UKGYF3PV5XNTCUL/action/replication_record"}},"created_at":"2026-07-05T02:51:38.850960+00:00","updated_at":"2026-07-05T02:51:38.850960+00:00"}