{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RIMIJCVGZYB52QC63LKU4FZ7A6","short_pith_number":"pith:RIMIJCVG","schema_version":"1.0","canonical_sha256":"8a18848aa6ce03dd405edad54e173f0789361eb093478f9c09a43852a568c953","source":{"kind":"arxiv","id":"2505.03273","version":2},"attestation_state":"computed","paper":{"title":"SepALM: Audio Language Models Are Error Correctors for Robust Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Gang Wang, Xinyu Yang, Zhaoxi Mu","submitted_at":"2025-05-06T08:04:37Z","abstract_excerpt":"While contemporary speech separation technologies adeptly process lengthy mixed audio waveforms, they are frequently challenged by the intricacies of real-world environments, including noisy and reverberant settings, which can result in artifacts or distortions in the separated speech. To overcome these limitations, we introduce SepALM, a pioneering approach that employs audio language models (ALMs) to rectify and re-synthesize speech within the text domain following preliminary separation. SepALM comprises four core components: a separator, a corrector, a synthesizer, and an aligner. By integ"},"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":"2505.03273","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-05-06T08:04:37Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"fc514d5d3ac5dc86b7b4b532c9f5df79672b66ae7a1e3fc6057a99f85f3c04cb","abstract_canon_sha256":"bc83ccf4ab9b9f48d021d3109bf039ae432175abfb98a017d67ded328beac0de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:12.675432Z","signature_b64":"UZrPlVR+biW7jeCw8TluHJmTVbl85yXP070n75ENeqmzFrBEOvE6LopU3ByA7h9ZiQbI8Ega/H1qk9eP21IXAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a18848aa6ce03dd405edad54e173f0789361eb093478f9c09a43852a568c953","last_reissued_at":"2026-07-05T11:09:12.674915Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:12.674915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SepALM: Audio Language Models Are Error Correctors for Robust Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Gang Wang, Xinyu Yang, Zhaoxi Mu","submitted_at":"2025-05-06T08:04:37Z","abstract_excerpt":"While contemporary speech separation technologies adeptly process lengthy mixed audio waveforms, they are frequently challenged by the intricacies of real-world environments, including noisy and reverberant settings, which can result in artifacts or distortions in the separated speech. To overcome these limitations, we introduce SepALM, a pioneering approach that employs audio language models (ALMs) to rectify and re-synthesize speech within the text domain following preliminary separation. SepALM comprises four core components: a separator, a corrector, a synthesizer, and an aligner. By integ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03273","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/2505.03273/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":"2505.03273","created_at":"2026-07-05T11:09:12.674975+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03273v2","created_at":"2026-07-05T11:09:12.674975+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03273","created_at":"2026-07-05T11:09:12.674975+00:00"},{"alias_kind":"pith_short_12","alias_value":"RIMIJCVGZYB5","created_at":"2026-07-05T11:09:12.674975+00:00"},{"alias_kind":"pith_short_16","alias_value":"RIMIJCVGZYB52QC6","created_at":"2026-07-05T11:09:12.674975+00:00"},{"alias_kind":"pith_short_8","alias_value":"RIMIJCVG","created_at":"2026-07-05T11:09:12.674975+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":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6","json":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6.json","graph_json":"https://pith.science/api/pith-number/RIMIJCVGZYB52QC63LKU4FZ7A6/graph.json","events_json":"https://pith.science/api/pith-number/RIMIJCVGZYB52QC63LKU4FZ7A6/events.json","paper":"https://pith.science/paper/RIMIJCVG"},"agent_actions":{"view_html":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6","download_json":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6.json","view_paper":"https://pith.science/paper/RIMIJCVG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03273&json=true","fetch_graph":"https://pith.science/api/pith-number/RIMIJCVGZYB52QC63LKU4FZ7A6/graph.json","fetch_events":"https://pith.science/api/pith-number/RIMIJCVGZYB52QC63LKU4FZ7A6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6/action/storage_attestation","attest_author":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6/action/author_attestation","sign_citation":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6/action/citation_signature","submit_replication":"https://pith.science/pith/RIMIJCVGZYB52QC63LKU4FZ7A6/action/replication_record"}},"created_at":"2026-07-05T11:09:12.674975+00:00","updated_at":"2026-07-05T11:09:12.674975+00:00"}