{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:FUT5DYSXV24I324PRSRF5HQDUU","short_pith_number":"pith:FUT5DYSX","schema_version":"1.0","canonical_sha256":"2d27d1e257aeb88deb8f8ca25e9e03a53bff4f56113fbd7ed5194db559c42901","source":{"kind":"arxiv","id":"1908.01768","version":1},"attestation_state":"computed","paper":{"title":"Probabilistic Permutation Invariant Training for Speech Separation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"John H.L. Hansen, Midia Yousefi, Soheil Khorram","submitted_at":"2019-08-04T17:42:31Z","abstract_excerpt":"Single-microphone, speaker-independent speech separation is normally performed through two steps: (i) separating the specific speech sources, and (ii) determining the best output-label assignment to find the separation error. The second step is the main obstacle in training neural networks for speech separation. Recently proposed Permutation Invariant Training (PIT) addresses this problem by determining the output-label assignment which minimizes the separation error. In this study, we show that a major drawback of this technique is the overconfident choice of the output-label assignment, espe"},"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":"1908.01768","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2019-08-04T17:42:31Z","cross_cats_sorted":["cs.LG","cs.SD","stat.ML"],"title_canon_sha256":"119fdb120774c111692547da45ec762859351e84d889d7c7772c0b3651cbcd41","abstract_canon_sha256":"8b3609758b9b5967bd17ad6843a7829b162ec0358ab0b642d1c023ef6d65b19f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:51:47.437919Z","signature_b64":"IhH9UDqR9iOdOveNi3bpVdCh3fUYHMeyrPO9CrQROCaDom3N1OJHfDTXwxLNu11ZHKlewfX/pvj8qoA9/t8TAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d27d1e257aeb88deb8f8ca25e9e03a53bff4f56113fbd7ed5194db559c42901","last_reissued_at":"2026-07-04T23:51:47.437582Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:51:47.437582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Probabilistic Permutation Invariant Training for Speech Separation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"John H.L. Hansen, Midia Yousefi, Soheil Khorram","submitted_at":"2019-08-04T17:42:31Z","abstract_excerpt":"Single-microphone, speaker-independent speech separation is normally performed through two steps: (i) separating the specific speech sources, and (ii) determining the best output-label assignment to find the separation error. The second step is the main obstacle in training neural networks for speech separation. Recently proposed Permutation Invariant Training (PIT) addresses this problem by determining the output-label assignment which minimizes the separation error. In this study, we show that a major drawback of this technique is the overconfident choice of the output-label assignment, espe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.01768","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/1908.01768/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":"1908.01768","created_at":"2026-07-04T23:51:47.437639+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.01768v1","created_at":"2026-07-04T23:51:47.437639+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.01768","created_at":"2026-07-04T23:51:47.437639+00:00"},{"alias_kind":"pith_short_12","alias_value":"FUT5DYSXV24I","created_at":"2026-07-04T23:51:47.437639+00:00"},{"alias_kind":"pith_short_16","alias_value":"FUT5DYSXV24I324P","created_at":"2026-07-04T23:51:47.437639+00:00"},{"alias_kind":"pith_short_8","alias_value":"FUT5DYSX","created_at":"2026-07-04T23:51:47.437639+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1908.01768","citing_title":"Probabilistic Permutation Invariant Training for Speech Separation","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU","json":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU.json","graph_json":"https://pith.science/api/pith-number/FUT5DYSXV24I324PRSRF5HQDUU/graph.json","events_json":"https://pith.science/api/pith-number/FUT5DYSXV24I324PRSRF5HQDUU/events.json","paper":"https://pith.science/paper/FUT5DYSX"},"agent_actions":{"view_html":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU","download_json":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU.json","view_paper":"https://pith.science/paper/FUT5DYSX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.01768&json=true","fetch_graph":"https://pith.science/api/pith-number/FUT5DYSXV24I324PRSRF5HQDUU/graph.json","fetch_events":"https://pith.science/api/pith-number/FUT5DYSXV24I324PRSRF5HQDUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU/action/storage_attestation","attest_author":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU/action/author_attestation","sign_citation":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU/action/citation_signature","submit_replication":"https://pith.science/pith/FUT5DYSXV24I324PRSRF5HQDUU/action/replication_record"}},"created_at":"2026-07-04T23:51:47.437639+00:00","updated_at":"2026-07-04T23:51:47.437639+00:00"}