{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EEEFKE6MBPTGJS5DJ7WJUDBNCY","short_pith_number":"pith:EEEFKE6M","schema_version":"1.0","canonical_sha256":"21085513cc0be664cba34fec9a0c2d1623739fb393a8616283bd731cf4d07392","source":{"kind":"arxiv","id":"1901.05061","version":3},"attestation_state":"computed","paper":{"title":"Spectrogram Feature Losses for Music Source Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS","stat.ML"],"primary_cat":"cs.SD","authors_text":"Abhimanyu Sahai, Brian McWilliams, Romann Weber","submitted_at":"2019-01-15T21:51:12Z","abstract_excerpt":"In this paper we study deep learning-based music source separation, and explore using an alternative loss to the standard spectrogram pixel-level L2 loss for model training. Our main contribution is in demonstrating that adding a high-level feature loss term, extracted from the spectrograms using a VGG net, can improve separation quality vis-a-vis a pure pixel-level loss. We show this improvement in the context of the MMDenseNet, a State-of-the-Art deep learning model for this task, for the extraction of drums and vocal sounds from songs in the musdb18 database, covering a broad range of weste"},"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":"1901.05061","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2019-01-15T21:51:12Z","cross_cats_sorted":["cs.LG","eess.AS","stat.ML"],"title_canon_sha256":"f21ff4e021ebe6fb10392be2c43d7ddd82fec5c84f2a99f73ef720896b73fd10","abstract_canon_sha256":"d8ecb142347f1173ec680bee06e5602eec0c45e172e62900723152e3928adeee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:08.386379Z","signature_b64":"RyJr0pOImQ5hhWjZqtMfBl/B9wndEhsAqkXgOt2XjO5T6BTq0gOvwqceJo2zg3Rfi1E8UuSnkwFnKI/jJZIaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21085513cc0be664cba34fec9a0c2d1623739fb393a8616283bd731cf4d07392","last_reissued_at":"2026-05-17T23:42:08.385779Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:08.385779Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spectrogram Feature Losses for Music Source Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS","stat.ML"],"primary_cat":"cs.SD","authors_text":"Abhimanyu Sahai, Brian McWilliams, Romann Weber","submitted_at":"2019-01-15T21:51:12Z","abstract_excerpt":"In this paper we study deep learning-based music source separation, and explore using an alternative loss to the standard spectrogram pixel-level L2 loss for model training. Our main contribution is in demonstrating that adding a high-level feature loss term, extracted from the spectrograms using a VGG net, can improve separation quality vis-a-vis a pure pixel-level loss. We show this improvement in the context of the MMDenseNet, a State-of-the-Art deep learning model for this task, for the extraction of drums and vocal sounds from songs in the musdb18 database, covering a broad range of weste"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.05061","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":""},"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":"1901.05061","created_at":"2026-05-17T23:42:08.385875+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.05061v3","created_at":"2026-05-17T23:42:08.385875+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.05061","created_at":"2026-05-17T23:42:08.385875+00:00"},{"alias_kind":"pith_short_12","alias_value":"EEEFKE6MBPTG","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"EEEFKE6MBPTGJS5D","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"EEEFKE6M","created_at":"2026-05-18T12:33:15.570797+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/EEEFKE6MBPTGJS5DJ7WJUDBNCY","json":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY.json","graph_json":"https://pith.science/api/pith-number/EEEFKE6MBPTGJS5DJ7WJUDBNCY/graph.json","events_json":"https://pith.science/api/pith-number/EEEFKE6MBPTGJS5DJ7WJUDBNCY/events.json","paper":"https://pith.science/paper/EEEFKE6M"},"agent_actions":{"view_html":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY","download_json":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY.json","view_paper":"https://pith.science/paper/EEEFKE6M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.05061&json=true","fetch_graph":"https://pith.science/api/pith-number/EEEFKE6MBPTGJS5DJ7WJUDBNCY/graph.json","fetch_events":"https://pith.science/api/pith-number/EEEFKE6MBPTGJS5DJ7WJUDBNCY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY/action/storage_attestation","attest_author":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY/action/author_attestation","sign_citation":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY/action/citation_signature","submit_replication":"https://pith.science/pith/EEEFKE6MBPTGJS5DJ7WJUDBNCY/action/replication_record"}},"created_at":"2026-05-17T23:42:08.385875+00:00","updated_at":"2026-05-17T23:42:08.385875+00:00"}