{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:SLOCGXVAX5S5MFVREE3UA2VPUK","short_pith_number":"pith:SLOCGXVA","schema_version":"1.0","canonical_sha256":"92dc235ea0bf65d616b12137406aafa2be374c4f20431a166c45b2dde5ff0b39","source":{"kind":"arxiv","id":"1910.09804","version":2},"attestation_state":"computed","paper":{"title":"Two-Step Sound Source Separation: Training on Learned Latent Targets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cem Subakan, Efthymios Tzinis, Paris Smaragdis, Shrikant Venkataramani, Zhepei Wang","submitted_at":"2019-10-22T07:49:21Z","abstract_excerpt":"In this paper, we propose a two-step training procedure for source separation via a deep neural network. In the first step we learn a transform (and it's inverse) to a latent space where masking-based separation performance using oracles is optimal. For the second step, we train a separation module that operates on the previously learned space. In order to do so, we also make use of a scale-invariant signal to distortion ratio (SI-SDR) loss function that works in the latent space, and we prove that it lower-bounds the SI-SDR in the time domain. We run various sound separation experiments that "},"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":"1910.09804","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-22T07:49:21Z","cross_cats_sorted":["cs.CL","cs.SD","eess.AS","stat.ML"],"title_canon_sha256":"5efbd97d095f4d319ffc3c9a93e4e1c030eb764f0f451890206799f054851646","abstract_canon_sha256":"da9386428e74276d9172f67e9ff00677b7b850d9c53828bafe84b09dcaeab28f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:38:59.931846Z","signature_b64":"AC3LbvTMjHAxFBbJnMGFaGI+eWu9aBSxQDJjc3GV0i1f0lTUcjJgvrLnA9x8EGyp8qp1sssL6JdwUwkD6knDAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"92dc235ea0bf65d616b12137406aafa2be374c4f20431a166c45b2dde5ff0b39","last_reissued_at":"2026-07-05T02:38:59.931402Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:38:59.931402Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Two-Step Sound Source Separation: Training on Learned Latent Targets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cem Subakan, Efthymios Tzinis, Paris Smaragdis, Shrikant Venkataramani, Zhepei Wang","submitted_at":"2019-10-22T07:49:21Z","abstract_excerpt":"In this paper, we propose a two-step training procedure for source separation via a deep neural network. In the first step we learn a transform (and it's inverse) to a latent space where masking-based separation performance using oracles is optimal. For the second step, we train a separation module that operates on the previously learned space. In order to do so, we also make use of a scale-invariant signal to distortion ratio (SI-SDR) loss function that works in the latent space, and we prove that it lower-bounds the SI-SDR in the time domain. We run various sound separation experiments that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.09804","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/1910.09804/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":"1910.09804","created_at":"2026-07-05T02:38:59.931459+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.09804v2","created_at":"2026-07-05T02:38:59.931459+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.09804","created_at":"2026-07-05T02:38:59.931459+00:00"},{"alias_kind":"pith_short_12","alias_value":"SLOCGXVAX5S5","created_at":"2026-07-05T02:38:59.931459+00:00"},{"alias_kind":"pith_short_16","alias_value":"SLOCGXVAX5S5MFVR","created_at":"2026-07-05T02:38:59.931459+00:00"},{"alias_kind":"pith_short_8","alias_value":"SLOCGXVA","created_at":"2026-07-05T02:38:59.931459+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK","json":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK.json","graph_json":"https://pith.science/api/pith-number/SLOCGXVAX5S5MFVREE3UA2VPUK/graph.json","events_json":"https://pith.science/api/pith-number/SLOCGXVAX5S5MFVREE3UA2VPUK/events.json","paper":"https://pith.science/paper/SLOCGXVA"},"agent_actions":{"view_html":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK","download_json":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK.json","view_paper":"https://pith.science/paper/SLOCGXVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.09804&json=true","fetch_graph":"https://pith.science/api/pith-number/SLOCGXVAX5S5MFVREE3UA2VPUK/graph.json","fetch_events":"https://pith.science/api/pith-number/SLOCGXVAX5S5MFVREE3UA2VPUK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK/action/storage_attestation","attest_author":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK/action/author_attestation","sign_citation":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK/action/citation_signature","submit_replication":"https://pith.science/pith/SLOCGXVAX5S5MFVREE3UA2VPUK/action/replication_record"}},"created_at":"2026-07-05T02:38:59.931459+00:00","updated_at":"2026-07-05T02:38:59.931459+00:00"}