{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SVTSCCEFZZQ7CTJQPSQSWPKHUU","short_pith_number":"pith:SVTSCCEF","schema_version":"1.0","canonical_sha256":"9567210885ce61f14d307ca12b3d47a50ce9e131c30944a46dd4446ba3586a15","source":{"kind":"arxiv","id":"2406.02092","version":1},"attestation_state":"computed","paper":{"title":"MaskSR: Masked Language Model for Full-band Speech Restoration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS","eess.SP"],"primary_cat":"cs.SD","authors_text":"Qirui Wang, Xiaoyu Liu, Xu Li","submitted_at":"2024-06-04T08:23:57Z","abstract_excerpt":"Speech restoration aims at restoring high quality speech in the presence of a diverse set of distortions. Although several deep learning paradigms have been studied for this task, the power of the recently emerging language models has not been fully explored. In this paper, we propose MaskSR, a masked language model capable of restoring full-band 44.1 kHz speech jointly considering noise, reverb, clipping, and low bandwidth. MaskSR works with discrete acoustic tokens extracted using a pre-trained neural codec. During training, MaskSR is optimized to predict randomly masked tokens extracted fro"},"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":"2406.02092","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-06-04T08:23:57Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS","eess.SP"],"title_canon_sha256":"1a19f03281ee9cf458472b9a06f00543efac54e914fe046a0c7de3362a404a78","abstract_canon_sha256":"c3a34d0d9588212e6dbaba0c542a9771f0c390d97d320064cf6bc5b374d02c38"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:14.592258Z","signature_b64":"FN3zxR5s/WUyc0v8f+yLO8bpma61pRBUupb8I8GFXeIOXbvc6RyLcCkZtJzSdqUEbcUYCeIjCEpwLQKfUxm6AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9567210885ce61f14d307ca12b3d47a50ce9e131c30944a46dd4446ba3586a15","last_reissued_at":"2026-07-05T08:27:14.591848Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:14.591848Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MaskSR: Masked Language Model for Full-band Speech Restoration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS","eess.SP"],"primary_cat":"cs.SD","authors_text":"Qirui Wang, Xiaoyu Liu, Xu Li","submitted_at":"2024-06-04T08:23:57Z","abstract_excerpt":"Speech restoration aims at restoring high quality speech in the presence of a diverse set of distortions. Although several deep learning paradigms have been studied for this task, the power of the recently emerging language models has not been fully explored. In this paper, we propose MaskSR, a masked language model capable of restoring full-band 44.1 kHz speech jointly considering noise, reverb, clipping, and low bandwidth. MaskSR works with discrete acoustic tokens extracted using a pre-trained neural codec. During training, MaskSR is optimized to predict randomly masked tokens extracted fro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02092","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/2406.02092/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":"2406.02092","created_at":"2026-07-05T08:27:14.591902+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.02092v1","created_at":"2026-07-05T08:27:14.591902+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02092","created_at":"2026-07-05T08:27:14.591902+00:00"},{"alias_kind":"pith_short_12","alias_value":"SVTSCCEFZZQ7","created_at":"2026-07-05T08:27:14.591902+00:00"},{"alias_kind":"pith_short_16","alias_value":"SVTSCCEFZZQ7CTJQ","created_at":"2026-07-05T08:27:14.591902+00:00"},{"alias_kind":"pith_short_8","alias_value":"SVTSCCEF","created_at":"2026-07-05T08:27:14.591902+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31247","citing_title":"FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model","ref_index":156,"is_internal_anchor":false},{"citing_arxiv_id":"2508.03448","citing_title":"SonicMaster: Towards Controllable All-in-One Music Restoration and Mastering","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU","json":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU.json","graph_json":"https://pith.science/api/pith-number/SVTSCCEFZZQ7CTJQPSQSWPKHUU/graph.json","events_json":"https://pith.science/api/pith-number/SVTSCCEFZZQ7CTJQPSQSWPKHUU/events.json","paper":"https://pith.science/paper/SVTSCCEF"},"agent_actions":{"view_html":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU","download_json":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU.json","view_paper":"https://pith.science/paper/SVTSCCEF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.02092&json=true","fetch_graph":"https://pith.science/api/pith-number/SVTSCCEFZZQ7CTJQPSQSWPKHUU/graph.json","fetch_events":"https://pith.science/api/pith-number/SVTSCCEFZZQ7CTJQPSQSWPKHUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU/action/storage_attestation","attest_author":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU/action/author_attestation","sign_citation":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU/action/citation_signature","submit_replication":"https://pith.science/pith/SVTSCCEFZZQ7CTJQPSQSWPKHUU/action/replication_record"}},"created_at":"2026-07-05T08:27:14.591902+00:00","updated_at":"2026-07-05T08:27:14.591902+00:00"}