{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LQIKWGXGZASO4ELDG5UPV7UTUZ","short_pith_number":"pith:LQIKWGXG","schema_version":"1.0","canonical_sha256":"5c10ab1ae6c824ee11633768fafe93a65aec18faac6abebc5ca4b586c06df315","source":{"kind":"arxiv","id":"2105.02446","version":6},"attestation_state":"computed","paper":{"title":"DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chengxi Li, Feiyang Chen, Jinglin Liu, Yi Ren, Zhou Zhao","submitted_at":"2021-05-06T05:21:42Z","abstract_excerpt":"Singing voice synthesis (SVS) systems are built to synthesize high-quality and expressive singing voice, in which the acoustic model generates the acoustic features (e.g., mel-spectrogram) given a music score. Previous singing acoustic models adopt a simple loss (e.g., L1 and L2) or generative adversarial network (GAN) to reconstruct the acoustic features, while they suffer from over-smoothing and unstable training issues respectively, which hinder the naturalness of synthesized singing. In this work, we propose DiffSinger, an acoustic model for SVS based on the diffusion probabilistic model. "},"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":"2105.02446","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2021-05-06T05:21:42Z","cross_cats_sorted":["cs.LG","cs.SD"],"title_canon_sha256":"154cd4ab7affaa85bf4c53043ab94a6e5bb09e7cb59b4be8521414202314423b","abstract_canon_sha256":"83fbc8da64f91014bb5be848fcb1ca7c27aa668f98697ee5127d9c9a8b6c5947"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:07:03.828671Z","signature_b64":"aW38P2CDB2rmB5ZQTlMX7Xuhnzf9KdtGvLquN2XIHn2ia06RMXPwDpnB+lgi5XP4erLqizSj0XZiFVxRIMyFAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c10ab1ae6c824ee11633768fafe93a65aec18faac6abebc5ca4b586c06df315","last_reissued_at":"2026-07-05T04:07:03.828232Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:07:03.828232Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Chengxi Li, Feiyang Chen, Jinglin Liu, Yi Ren, Zhou Zhao","submitted_at":"2021-05-06T05:21:42Z","abstract_excerpt":"Singing voice synthesis (SVS) systems are built to synthesize high-quality and expressive singing voice, in which the acoustic model generates the acoustic features (e.g., mel-spectrogram) given a music score. Previous singing acoustic models adopt a simple loss (e.g., L1 and L2) or generative adversarial network (GAN) to reconstruct the acoustic features, while they suffer from over-smoothing and unstable training issues respectively, which hinder the naturalness of synthesized singing. In this work, we propose DiffSinger, an acoustic model for SVS based on the diffusion probabilistic model. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.02446","kind":"arxiv","version":6},"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/2105.02446/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":"2105.02446","created_at":"2026-07-05T04:07:03.828293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.02446v6","created_at":"2026-07-05T04:07:03.828293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.02446","created_at":"2026-07-05T04:07:03.828293+00:00"},{"alias_kind":"pith_short_12","alias_value":"LQIKWGXGZASO","created_at":"2026-07-05T04:07:03.828293+00:00"},{"alias_kind":"pith_short_16","alias_value":"LQIKWGXGZASO4ELD","created_at":"2026-07-05T04:07:03.828293+00:00"},{"alias_kind":"pith_short_8","alias_value":"LQIKWGXG","created_at":"2026-07-05T04:07:03.828293+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.04841","citing_title":"Joint Fullband-Subband Modeling for High-Resolution SingFake Detection","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ","json":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ.json","graph_json":"https://pith.science/api/pith-number/LQIKWGXGZASO4ELDG5UPV7UTUZ/graph.json","events_json":"https://pith.science/api/pith-number/LQIKWGXGZASO4ELDG5UPV7UTUZ/events.json","paper":"https://pith.science/paper/LQIKWGXG"},"agent_actions":{"view_html":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ","download_json":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ.json","view_paper":"https://pith.science/paper/LQIKWGXG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.02446&json=true","fetch_graph":"https://pith.science/api/pith-number/LQIKWGXGZASO4ELDG5UPV7UTUZ/graph.json","fetch_events":"https://pith.science/api/pith-number/LQIKWGXGZASO4ELDG5UPV7UTUZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ/action/storage_attestation","attest_author":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ/action/author_attestation","sign_citation":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ/action/citation_signature","submit_replication":"https://pith.science/pith/LQIKWGXGZASO4ELDG5UPV7UTUZ/action/replication_record"}},"created_at":"2026-07-05T04:07:03.828293+00:00","updated_at":"2026-07-05T04:07:03.828293+00:00"}