{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:L7NFRXV43F2C2PFQJNRKTS2G44","short_pith_number":"pith:L7NFRXV4","schema_version":"1.0","canonical_sha256":"5fda58debcd9742d3cb04b62a9cb46e73abedf960476526658157554f60a0f1d","source":{"kind":"arxiv","id":"2106.07843","version":3},"attestation_state":"computed","paper":{"title":"Teacher-Student MixIT for Unsupervised and Semi-supervised Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Catalin Zorila, Jisi Zhang, Jon Barker, Rama Doddipatla","submitted_at":"2021-06-15T02:26:42Z","abstract_excerpt":"In this paper, we introduce a novel semi-supervised learning framework for end-to-end speech separation. The proposed method first uses mixtures of unseparated sources and the mixture invariant training (MixIT) criterion to train a teacher model. The teacher model then estimates separated sources that are used to train a student model with standard permutation invariant training (PIT). The student model can be fine-tuned with supervised data, i.e., paired artificial mixtures and clean speech sources, and further improved via model distillation. Experiments with single and multi channel mixture"},"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":"2106.07843","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2021-06-15T02:26:42Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"44b3e07ccda61e6ff6826f0ea0ba53b737029626171a7c3f68578c30e6758ade","abstract_canon_sha256":"5cb95f1a94c7034bcb4b77bdf356256ba02253246f771f26f12ad9ca4447fbc3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:12:49.296973Z","signature_b64":"TDDPUn1DjJedl1/GeRidLEaPMogRtI4UJyIn0Z4ExoJt4i0m3p0avJUHVwH3gAvur6Cy/4Mm88L9UTE7r495Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fda58debcd9742d3cb04b62a9cb46e73abedf960476526658157554f60a0f1d","last_reissued_at":"2026-07-05T03:12:49.296505Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:12:49.296505Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Teacher-Student MixIT for Unsupervised and Semi-supervised Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Catalin Zorila, Jisi Zhang, Jon Barker, Rama Doddipatla","submitted_at":"2021-06-15T02:26:42Z","abstract_excerpt":"In this paper, we introduce a novel semi-supervised learning framework for end-to-end speech separation. The proposed method first uses mixtures of unseparated sources and the mixture invariant training (MixIT) criterion to train a teacher model. The teacher model then estimates separated sources that are used to train a student model with standard permutation invariant training (PIT). The student model can be fine-tuned with supervised data, i.e., paired artificial mixtures and clean speech sources, and further improved via model distillation. Experiments with single and multi channel mixture"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.07843","kind":"arxiv","version":3},"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/2106.07843/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":"2106.07843","created_at":"2026-07-05T03:12:49.296569+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.07843v3","created_at":"2026-07-05T03:12:49.296569+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.07843","created_at":"2026-07-05T03:12:49.296569+00:00"},{"alias_kind":"pith_short_12","alias_value":"L7NFRXV43F2C","created_at":"2026-07-05T03:12:49.296569+00:00"},{"alias_kind":"pith_short_16","alias_value":"L7NFRXV43F2C2PFQ","created_at":"2026-07-05T03:12:49.296569+00:00"},{"alias_kind":"pith_short_8","alias_value":"L7NFRXV4","created_at":"2026-07-05T03:12:49.296569+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.10830","citing_title":"Advances in Speech Separation: Techniques, Challenges, and Future Trends","ref_index":82,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44","json":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44.json","graph_json":"https://pith.science/api/pith-number/L7NFRXV43F2C2PFQJNRKTS2G44/graph.json","events_json":"https://pith.science/api/pith-number/L7NFRXV43F2C2PFQJNRKTS2G44/events.json","paper":"https://pith.science/paper/L7NFRXV4"},"agent_actions":{"view_html":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44","download_json":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44.json","view_paper":"https://pith.science/paper/L7NFRXV4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.07843&json=true","fetch_graph":"https://pith.science/api/pith-number/L7NFRXV43F2C2PFQJNRKTS2G44/graph.json","fetch_events":"https://pith.science/api/pith-number/L7NFRXV43F2C2PFQJNRKTS2G44/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44/action/storage_attestation","attest_author":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44/action/author_attestation","sign_citation":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44/action/citation_signature","submit_replication":"https://pith.science/pith/L7NFRXV43F2C2PFQJNRKTS2G44/action/replication_record"}},"created_at":"2026-07-05T03:12:49.296569+00:00","updated_at":"2026-07-05T03:12:49.296569+00:00"}