{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RIRVDSDY4YB35YJXHXBUOKMQQV","short_pith_number":"pith:RIRVDSDY","schema_version":"1.0","canonical_sha256":"8a2351c878e603bee1373dc3472990857239539d4c55a98f5e7a75bd6c1a7447","source":{"kind":"arxiv","id":"2409.16117","version":2},"attestation_state":"computed","paper":{"title":"Generative Speech Foundation Model Pretraining for High-Quality Speech Extraction and Restoration","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Alexander H. Liu, Ante Juki\\'c, Pin-Jui Ku, Roman Korostik, Sung-Feng Huang, Szu-Wei Fu","submitted_at":"2024-09-24T14:24:47Z","abstract_excerpt":"This paper proposes a generative pretraining foundation model for high-quality speech restoration tasks. By directly operating on complex-valued short-time Fourier transform coefficients, our model does not rely on any vocoders for time-domain signal reconstruction. As a result, our model simplifies the synthesis process and removes the quality upper-bound introduced by any mel-spectrogram vocoder compared to prior work SpeechFlow. The proposed method is evaluated on multiple speech restoration tasks, including speech denoising, bandwidth extension, codec artifact removal, and target speaker e"},"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":"2409.16117","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2024-09-24T14:24:47Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"76ea9fab263dcc1c6c66012a5f3273872d9b8b056ace04d12e3dcfdf4a5cc22e","abstract_canon_sha256":"a08f2b91011d9602616984ac5c66b94b43af9b2f9e07f17210bdbd55b1770ad7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:11:30.847560Z","signature_b64":"kTPPRyLTjSFEWRcpmwEBiPQhFqhvFLE5y49kHgqzrlMBnXe4d3YbMFqQEog5jbYkcyvsJ+TYTavu88IJTeuWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a2351c878e603bee1373dc3472990857239539d4c55a98f5e7a75bd6c1a7447","last_reissued_at":"2026-07-05T09:11:30.847110Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:11:30.847110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Speech Foundation Model Pretraining for High-Quality Speech Extraction and Restoration","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Alexander H. Liu, Ante Juki\\'c, Pin-Jui Ku, Roman Korostik, Sung-Feng Huang, Szu-Wei Fu","submitted_at":"2024-09-24T14:24:47Z","abstract_excerpt":"This paper proposes a generative pretraining foundation model for high-quality speech restoration tasks. By directly operating on complex-valued short-time Fourier transform coefficients, our model does not rely on any vocoders for time-domain signal reconstruction. As a result, our model simplifies the synthesis process and removes the quality upper-bound introduced by any mel-spectrogram vocoder compared to prior work SpeechFlow. The proposed method is evaluated on multiple speech restoration tasks, including speech denoising, bandwidth extension, codec artifact removal, and target speaker e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.16117","kind":"arxiv","version":2},"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/2409.16117/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":"2409.16117","created_at":"2026-07-05T09:11:30.847167+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.16117v2","created_at":"2026-07-05T09:11:30.847167+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.16117","created_at":"2026-07-05T09:11:30.847167+00:00"},{"alias_kind":"pith_short_12","alias_value":"RIRVDSDY4YB3","created_at":"2026-07-05T09:11:30.847167+00:00"},{"alias_kind":"pith_short_16","alias_value":"RIRVDSDY4YB35YJX","created_at":"2026-07-05T09:11:30.847167+00:00"},{"alias_kind":"pith_short_8","alias_value":"RIRVDSDY","created_at":"2026-07-05T09:11:30.847167+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.19314","citing_title":"SoloSpeech: Enhancing Intelligibility and Quality in Target Speech Extraction through a Cascaded Generative Pipeline","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV","json":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV.json","graph_json":"https://pith.science/api/pith-number/RIRVDSDY4YB35YJXHXBUOKMQQV/graph.json","events_json":"https://pith.science/api/pith-number/RIRVDSDY4YB35YJXHXBUOKMQQV/events.json","paper":"https://pith.science/paper/RIRVDSDY"},"agent_actions":{"view_html":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV","download_json":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV.json","view_paper":"https://pith.science/paper/RIRVDSDY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.16117&json=true","fetch_graph":"https://pith.science/api/pith-number/RIRVDSDY4YB35YJXHXBUOKMQQV/graph.json","fetch_events":"https://pith.science/api/pith-number/RIRVDSDY4YB35YJXHXBUOKMQQV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV/action/storage_attestation","attest_author":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV/action/author_attestation","sign_citation":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV/action/citation_signature","submit_replication":"https://pith.science/pith/RIRVDSDY4YB35YJXHXBUOKMQQV/action/replication_record"}},"created_at":"2026-07-05T09:11:30.847167+00:00","updated_at":"2026-07-05T09:11:30.847167+00:00"}