{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:KNCTHVE6JE7YDYMZFBYK7MLPK3","short_pith_number":"pith:KNCTHVE6","schema_version":"1.0","canonical_sha256":"534533d49e493f81e1992870afb16f56ef474dbb43733e1e87071739e8a8ce14","source":{"kind":"arxiv","id":"2005.11262","version":1},"attestation_state":"computed","paper":{"title":"LibriMix: An Open-Source Dataset for Generalizable Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Antoine Deleforge, Emmanuel Vincent, Joris Cosentino, Manuel Pariente, Samuele Cornell","submitted_at":"2020-05-22T16:26:54Z","abstract_excerpt":"In recent years, wsj0-2mix has become the reference dataset for single-channel speech separation. Most deep learning-based speech separation models today are benchmarked on it. However, recent studies have shown important performance drops when models trained on wsj0-2mix are evaluated on other, similar datasets. To address this generalization issue, we created LibriMix, an open-source alternative to wsj0-2mix, and to its noisy extension, WHAM!. Based on LibriSpeech, LibriMix consists of two- or three-speaker mixtures combined with ambient noise samples from WHAM!. Using Conv-TasNet, we achiev"},"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":"2005.11262","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-05-22T16:26:54Z","cross_cats_sorted":[],"title_canon_sha256":"30ee8dddc68a9ec20246046c41b197703646683d290ff36418967ea053c35d49","abstract_canon_sha256":"80587a6b982f02fe0fdf6d8cd151023e0fb38125f7a6691132ebb469b6cb00a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:05:06.672930Z","signature_b64":"QJqZa6iQPFHgontuvo2m0o6PK5drN/6Azcc3N0Gfyg1ea5ATBf1G1zWqMCeexBs0zD05PeZQdq/3MwSewAyCDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"534533d49e493f81e1992870afb16f56ef474dbb43733e1e87071739e8a8ce14","last_reissued_at":"2026-07-05T01:05:06.672506Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:05:06.672506Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LibriMix: An Open-Source Dataset for Generalizable Speech Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Antoine Deleforge, Emmanuel Vincent, Joris Cosentino, Manuel Pariente, Samuele Cornell","submitted_at":"2020-05-22T16:26:54Z","abstract_excerpt":"In recent years, wsj0-2mix has become the reference dataset for single-channel speech separation. Most deep learning-based speech separation models today are benchmarked on it. However, recent studies have shown important performance drops when models trained on wsj0-2mix are evaluated on other, similar datasets. To address this generalization issue, we created LibriMix, an open-source alternative to wsj0-2mix, and to its noisy extension, WHAM!. Based on LibriSpeech, LibriMix consists of two- or three-speaker mixtures combined with ambient noise samples from WHAM!. Using Conv-TasNet, we achiev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.11262","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/2005.11262/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":"2005.11262","created_at":"2026-07-05T01:05:06.672566+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.11262v1","created_at":"2026-07-05T01:05:06.672566+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.11262","created_at":"2026-07-05T01:05:06.672566+00:00"},{"alias_kind":"pith_short_12","alias_value":"KNCTHVE6JE7Y","created_at":"2026-07-05T01:05:06.672566+00:00"},{"alias_kind":"pith_short_16","alias_value":"KNCTHVE6JE7YDYMZ","created_at":"2026-07-05T01:05:06.672566+00:00"},{"alias_kind":"pith_short_8","alias_value":"KNCTHVE6","created_at":"2026-07-05T01:05:06.672566+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.08111","citing_title":"PS4: Proxy-Supervised Joint Training for Real Target Speaker Extraction","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2607.06088","citing_title":"Flow Matching-Based Speech Source Separation with Best-of-N Biometric Sampling","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23313","citing_title":"Uncertainty-based Debiasing and Unlearning for Decontamination","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23332","citing_title":"Don't Listen to Me: A Lightweight, Low-Latency Model for Own-Voice Cancellation in Far-Field Speech Enhancement","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29897","citing_title":"Child-Centric Voice Anonymization in Single and Multi-Speaker Speech via Domain-Adapted SSL Models","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2406.14294","citing_title":"DASB - Discrete Audio and Speech Benchmark","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2601.06006","citing_title":"Discriminative-Generative Target Speaker Extraction with Decoder-Only Language Models","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17225","citing_title":"Can Large Audio Language Models Ignore Multilingual Distractors? An Evaluation of Their Selective Auditory Attention Capabilities","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15442","citing_title":"Mind the Gap: Impact of Synthetic Conversational Data on Multi-Talker ASR and Speaker Diarization","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2310.13289","citing_title":"SALMONN: Towards Generic Hearing Abilities for Large Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11192","citing_title":"Exploring Token-Space Manipulation in Latent Audio Tokenizers","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27436","citing_title":"BUT System Description for CHiME-9 MCoRec Challenge","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08415","citing_title":"Ring Mixing with Auxiliary Signal-to-Consistency-Error Ratio Loss for Unsupervised Denoising in Speech Separation","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3","json":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3.json","graph_json":"https://pith.science/api/pith-number/KNCTHVE6JE7YDYMZFBYK7MLPK3/graph.json","events_json":"https://pith.science/api/pith-number/KNCTHVE6JE7YDYMZFBYK7MLPK3/events.json","paper":"https://pith.science/paper/KNCTHVE6"},"agent_actions":{"view_html":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3","download_json":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3.json","view_paper":"https://pith.science/paper/KNCTHVE6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.11262&json=true","fetch_graph":"https://pith.science/api/pith-number/KNCTHVE6JE7YDYMZFBYK7MLPK3/graph.json","fetch_events":"https://pith.science/api/pith-number/KNCTHVE6JE7YDYMZFBYK7MLPK3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3/action/storage_attestation","attest_author":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3/action/author_attestation","sign_citation":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3/action/citation_signature","submit_replication":"https://pith.science/pith/KNCTHVE6JE7YDYMZFBYK7MLPK3/action/replication_record"}},"created_at":"2026-07-05T01:05:06.672566+00:00","updated_at":"2026-07-05T01:05:06.672566+00:00"}