{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V5DZLA2AAGH3PRIGASHKNVVGV6","short_pith_number":"pith:V5DZLA2A","schema_version":"1.0","canonical_sha256":"af47958340018fb7c506048ea6d6a6af9bb626ba92210abb89bb3f9f7b47b88a","source":{"kind":"arxiv","id":"2506.11514","version":1},"attestation_state":"computed","paper":{"title":"Efficient Speech Enhancement via Embeddings from Pre-trained Generative Audioencoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Heinrich Dinkel, Jian Luan, Junbo Zhang, Linzhang Wang, Xingwei Sun, Yadong Niu","submitted_at":"2025-06-13T07:15:41Z","abstract_excerpt":"Recent research has delved into speech enhancement (SE) approaches that leverage audio embeddings from pre-trained models, diverging from time-frequency masking or signal prediction techniques. This paper introduces an efficient and extensible SE method. Our approach involves initially extracting audio embeddings from noisy speech using a pre-trained audioencoder, which are then denoised by a compact encoder network. Subsequently, a vocoder synthesizes the clean speech from denoised embeddings. An ablation study substantiates the parameter efficiency of the denoise encoder with a pre-trained a"},"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":"2506.11514","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-06-13T07:15:41Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"3bb51e164e232b8a509106c13865f4445950b030627eedc13d420c6aeade1eec","abstract_canon_sha256":"f49b2139f1e04c3dc3bd429c1d3dcd08322a59bef9cf49ccff6fb90fd393c78e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:05.870376Z","signature_b64":"a0/U4tPryVf5xJBqJ5YO0ag7mDHEAFwpmL7tKbzt4a7BISFPKohuT9Fgf3wTWfLMGtlVnTvJF4RgDP029JrqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af47958340018fb7c506048ea6d6a6af9bb626ba92210abb89bb3f9f7b47b88a","last_reissued_at":"2026-07-05T11:21:05.869860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:05.869860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Speech Enhancement via Embeddings from Pre-trained Generative Audioencoders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Heinrich Dinkel, Jian Luan, Junbo Zhang, Linzhang Wang, Xingwei Sun, Yadong Niu","submitted_at":"2025-06-13T07:15:41Z","abstract_excerpt":"Recent research has delved into speech enhancement (SE) approaches that leverage audio embeddings from pre-trained models, diverging from time-frequency masking or signal prediction techniques. This paper introduces an efficient and extensible SE method. Our approach involves initially extracting audio embeddings from noisy speech using a pre-trained audioencoder, which are then denoised by a compact encoder network. Subsequently, a vocoder synthesizes the clean speech from denoised embeddings. An ablation study substantiates the parameter efficiency of the denoise encoder with a pre-trained a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11514","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/2506.11514/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":"2506.11514","created_at":"2026-07-05T11:21:05.869926+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11514v1","created_at":"2026-07-05T11:21:05.869926+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11514","created_at":"2026-07-05T11:21:05.869926+00:00"},{"alias_kind":"pith_short_12","alias_value":"V5DZLA2AAGH3","created_at":"2026-07-05T11:21:05.869926+00:00"},{"alias_kind":"pith_short_16","alias_value":"V5DZLA2AAGH3PRIG","created_at":"2026-07-05T11:21:05.869926+00:00"},{"alias_kind":"pith_short_8","alias_value":"V5DZLA2A","created_at":"2026-07-05T11:21:05.869926+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.14606","citing_title":"UniPASE: A Generative Model for Universal Speech Enhancement with High Fidelity and Low Hallucinations","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6","json":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6.json","graph_json":"https://pith.science/api/pith-number/V5DZLA2AAGH3PRIGASHKNVVGV6/graph.json","events_json":"https://pith.science/api/pith-number/V5DZLA2AAGH3PRIGASHKNVVGV6/events.json","paper":"https://pith.science/paper/V5DZLA2A"},"agent_actions":{"view_html":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6","download_json":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6.json","view_paper":"https://pith.science/paper/V5DZLA2A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11514&json=true","fetch_graph":"https://pith.science/api/pith-number/V5DZLA2AAGH3PRIGASHKNVVGV6/graph.json","fetch_events":"https://pith.science/api/pith-number/V5DZLA2AAGH3PRIGASHKNVVGV6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6/action/storage_attestation","attest_author":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6/action/author_attestation","sign_citation":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6/action/citation_signature","submit_replication":"https://pith.science/pith/V5DZLA2AAGH3PRIGASHKNVVGV6/action/replication_record"}},"created_at":"2026-07-05T11:21:05.869926+00:00","updated_at":"2026-07-05T11:21:05.869926+00:00"}