{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MWAHC22OFHWSH4JOMLCCKL2F3V","short_pith_number":"pith:MWAHC22O","schema_version":"1.0","canonical_sha256":"6580716b4e29ed23f12e62c4252f45dd7beb41ba46998d01ee3a91e6183f4de9","source":{"kind":"arxiv","id":"2505.19476","version":2},"attestation_state":"computed","paper":{"title":"FlowSE: Efficient and High-Quality Speech Enhancement via Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.SP"],"primary_cat":"eess.AS","authors_text":"Chao Weng, Jun Chen, Lei Xie, Longshuai Xiao, Mingshuai Liu, Xinfa Zhu, Yike Zhu, Zikai Liu, Ziqian Wang","submitted_at":"2025-05-26T03:55:00Z","abstract_excerpt":"Generative models have excelled in audio tasks using approaches such as language models, diffusion, and flow matching. However, existing generative approaches for speech enhancement (SE) face notable challenges: language model-based methods suffer from quantization loss, leading to compromised speaker similarity and intelligibility, while diffusion models require complex training and high inference latency. To address these challenges, we propose FlowSE, a flow-matching-based model for SE. Flow matching learns a continuous transformation between noisy and clean speech distributions in a single"},"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":"2505.19476","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-05-26T03:55:00Z","cross_cats_sorted":["eess.SP"],"title_canon_sha256":"62cbaaaee76c36a557f55ea65b7ada2e2e5ffd7b69f65e3eaa10f942763c2bd2","abstract_canon_sha256":"ef0e672463b551693d708eb4abf65653facb863a61c86a48bfd6dac571733aea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:10.047453Z","signature_b64":"BumElUscSo+lwmTYckKnFcNJ5yve7ALkFJGq9NF0nHnVFYjGL+MpsnRujG3xrG1PNSPxYxciNuT7pWw9Eh92Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6580716b4e29ed23f12e62c4252f45dd7beb41ba46998d01ee3a91e6183f4de9","last_reissued_at":"2026-07-05T11:10:10.046971Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:10.046971Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FlowSE: Efficient and High-Quality Speech Enhancement via Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.SP"],"primary_cat":"eess.AS","authors_text":"Chao Weng, Jun Chen, Lei Xie, Longshuai Xiao, Mingshuai Liu, Xinfa Zhu, Yike Zhu, Zikai Liu, Ziqian Wang","submitted_at":"2025-05-26T03:55:00Z","abstract_excerpt":"Generative models have excelled in audio tasks using approaches such as language models, diffusion, and flow matching. However, existing generative approaches for speech enhancement (SE) face notable challenges: language model-based methods suffer from quantization loss, leading to compromised speaker similarity and intelligibility, while diffusion models require complex training and high inference latency. To address these challenges, we propose FlowSE, a flow-matching-based model for SE. Flow matching learns a continuous transformation between noisy and clean speech distributions in a single"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19476","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/2505.19476/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":"2505.19476","created_at":"2026-07-05T11:10:10.047025+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.19476v2","created_at":"2026-07-05T11:10:10.047025+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19476","created_at":"2026-07-05T11:10:10.047025+00:00"},{"alias_kind":"pith_short_12","alias_value":"MWAHC22OFHWS","created_at":"2026-07-05T11:10:10.047025+00:00"},{"alias_kind":"pith_short_16","alias_value":"MWAHC22OFHWSH4JO","created_at":"2026-07-05T11:10:10.047025+00:00"},{"alias_kind":"pith_short_8","alias_value":"MWAHC22O","created_at":"2026-07-05T11:10:10.047025+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.07558","citing_title":"UniFlow: Unifying Speech Front-End Tasks via Continuous Generative Modeling","ref_index":55,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V","json":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V.json","graph_json":"https://pith.science/api/pith-number/MWAHC22OFHWSH4JOMLCCKL2F3V/graph.json","events_json":"https://pith.science/api/pith-number/MWAHC22OFHWSH4JOMLCCKL2F3V/events.json","paper":"https://pith.science/paper/MWAHC22O"},"agent_actions":{"view_html":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V","download_json":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V.json","view_paper":"https://pith.science/paper/MWAHC22O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.19476&json=true","fetch_graph":"https://pith.science/api/pith-number/MWAHC22OFHWSH4JOMLCCKL2F3V/graph.json","fetch_events":"https://pith.science/api/pith-number/MWAHC22OFHWSH4JOMLCCKL2F3V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V/action/storage_attestation","attest_author":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V/action/author_attestation","sign_citation":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V/action/citation_signature","submit_replication":"https://pith.science/pith/MWAHC22OFHWSH4JOMLCCKL2F3V/action/replication_record"}},"created_at":"2026-07-05T11:10:10.047025+00:00","updated_at":"2026-07-05T11:10:10.047025+00:00"}