{"paper":{"title":"A Survey of Advancing Audio Super-Resolution and Bandwidth Extension from Discriminative to Generative Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Audio super-resolution is shifting from deterministic neural mappings that over-smooth high frequencies to generative models that sample plausible missing content.","cross_cats":["eess.SP"],"primary_cat":"eess.AS","authors_text":"Andrew C. Singer, Diego A. Cuji, Ningyuan Yang, Pu Zhao, Ryan M. Corey, Xue Lin, Yize Li","submitted_at":"2026-05-15T22:34:52Z","abstract_excerpt":"Audio super-resolution (SR), also referred to as bandwidth extension (BWE), aims to reconstruct high-fidelity signals from low-resolution (LR) or band-limited (BL) observations, an inherently ill-posed task due to the ambiguity of missing high-frequency (HF) content. This survey provides a comprehensive overview of the field, with a particular focus on the paradigm shift from discriminative mapping to modern generative modeling. We first review early discriminative deep neural network (DNN) models, which formulate BWE/SR as a deterministic mapping problem and are prone to regression-to-the-mea"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By providing a structured taxonomy and unified perspective, this survey establishes a comprehensive foundation and offers a practical roadmap for advancing BWE/SR from deterministic point estimation toward distribution-aware generative modeling.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The survey assumes that its selection of literature and proposed taxonomy accurately and comprehensively capture the key developments and trade-offs in the field without significant omissions or bias in coverage.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A structured survey of audio bandwidth extension that organizes the transition from deterministic discriminative DNNs to generative approaches including GANs, diffusion models, and flow-based methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Audio super-resolution is shifting from deterministic neural mappings that over-smooth high frequencies to generative models that sample plausible missing content.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"be7e0d05f0ce19d408303c1c0cee96a45e9a8228b224f3f8776bb11708697bbd"},"source":{"id":"2605.16681","kind":"arxiv","version":1},"verdict":{"id":"4af9e3da-d234-4070-93d6-0f1f67bf0760","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:23:18.122031Z","strongest_claim":"By providing a structured taxonomy and unified perspective, this survey establishes a comprehensive foundation and offers a practical roadmap for advancing BWE/SR from deterministic point estimation toward distribution-aware generative modeling.","one_line_summary":"A structured survey of audio bandwidth extension that organizes the transition from deterministic discriminative DNNs to generative approaches including GANs, diffusion models, and flow-based methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The survey assumes that its selection of literature and proposed taxonomy accurately and comprehensively capture the key developments and trade-offs in the field without significant omissions or bias in coverage.","pith_extraction_headline":"Audio super-resolution is shifting from deterministic neural mappings that over-smooth high frequencies to generative models that sample plausible missing content."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16681/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T20:32:32.102110Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T20:31:19.218939Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.381839Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.501808Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"bf2c24535da8257aa76d0c1df38749e684a010ebbef0eb8b065d536007338a25"},"references":{"count":72,"sample":[{"doi":"","year":null,"title":"Hifi++: a unified framework for neural vocoding, bandwidth extension and speech enhancement.arXiv preprint arXiv:2203.13086","work_id":"11b9b8ae-8e54-44e1-a39f-9c6fd06e2590","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling","work_id":"04430f4e-b270-479c-9dcd-bee723164789","ref_index":2,"cited_arxiv_id":"1803.01271","is_internal_anchor":true},{"doi":"","year":null,"title":"Hi-fi multi- speaker english tts dataset","work_id":"33ac3337-53e9-40fb-8848-68b9bfea3886","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Frequency-domain enhanced extreme bandwidth extension network with iccrn for superior speech quality","work_id":"aa737bee-014c-431b-87cd-665adf7ba0ce","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"CMGAN: Conformer- based metric gan for speech enhancement","work_id":"b1ebabc7-30bb-4e50-bc2d-9c54151d4fa5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":72,"snapshot_sha256":"ba620ea9eaeb9a2b7cfd4e8c1d863475f923d6805147191af419c93903765935","internal_anchors":20},"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"}