{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OVVGBTUGBGUZGHQYVR5TBYUA3Q","short_pith_number":"pith:OVVGBTUG","schema_version":"1.0","canonical_sha256":"756a60ce8609a9931e18ac7b30e280dc3051bca322cb0dcbb62e121a3fe17608","source":{"kind":"arxiv","id":"2608.03021","version":1},"attestation_state":"computed","paper":{"title":"MeloCodec: Harnessing Melodic Priors for High-Fidelity Singing Voice Representation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Chunfeng Wang, Hao Li, Kecan Mao, Qifei Li, Ruimin Wang, Wei Chen, Wenxin Fu, Ya Li, Yingming Gao, Yizhong Geng","submitted_at":"2026-08-04T02:06:46Z","abstract_excerpt":"Neural audio codecs serve as fundamental tokenizers for LLM-based audio generation. While semantic priors are widely exploited to enhance linguistic intelligibility, the integration of explicit acoustic priors remains underexplored, limiting synthesis fidelity in frequency-sensitive domains. To address this gap, we introduce MeloCodec, a novel framework designed to effectively incorporate melodic priors, a critical form of acoustic information for singing. To address the optimization instability typically caused by the direct fusion of such explicit priors, we propose a Tokenize-then-Fuse para"},"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":"2608.03021","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-08-04T02:06:46Z","cross_cats_sorted":[],"title_canon_sha256":"a7b9a78d071e31e56389d79f4a2f506642ad57ec4998f0763d7d37e1af6e70e0","abstract_canon_sha256":"9040f36f269f0f2bf6b41a088310bbc7290e02f4d9971003a64702e171c73ee8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:44:37.739840Z","signature_b64":"E9kP+P5ri5nY1oJMwVbGsyG1FBzK5EGGFNh34Gumi7pJwptMWzBx6Cp0I0CAIX83YUyu2hLvybUI1/cNqLIrCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"756a60ce8609a9931e18ac7b30e280dc3051bca322cb0dcbb62e121a3fe17608","last_reissued_at":"2026-08-05T00:44:37.737154Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:44:37.737154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MeloCodec: Harnessing Melodic Priors for High-Fidelity Singing Voice Representation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Chunfeng Wang, Hao Li, Kecan Mao, Qifei Li, Ruimin Wang, Wei Chen, Wenxin Fu, Ya Li, Yingming Gao, Yizhong Geng","submitted_at":"2026-08-04T02:06:46Z","abstract_excerpt":"Neural audio codecs serve as fundamental tokenizers for LLM-based audio generation. While semantic priors are widely exploited to enhance linguistic intelligibility, the integration of explicit acoustic priors remains underexplored, limiting synthesis fidelity in frequency-sensitive domains. To address this gap, we introduce MeloCodec, a novel framework designed to effectively incorporate melodic priors, a critical form of acoustic information for singing. To address the optimization instability typically caused by the direct fusion of such explicit priors, we propose a Tokenize-then-Fuse para"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03021","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/2608.03021/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":"2608.03021","created_at":"2026-08-05T00:44:37.738103+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.03021v1","created_at":"2026-08-05T00:44:37.738103+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03021","created_at":"2026-08-05T00:44:37.738103+00:00"},{"alias_kind":"pith_short_12","alias_value":"OVVGBTUGBGUZ","created_at":"2026-08-05T00:44:37.738103+00:00"},{"alias_kind":"pith_short_16","alias_value":"OVVGBTUGBGUZGHQY","created_at":"2026-08-05T00:44:37.738103+00:00"},{"alias_kind":"pith_short_8","alias_value":"OVVGBTUG","created_at":"2026-08-05T00:44:37.738103+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/OVVGBTUGBGUZGHQYVR5TBYUA3Q","json":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q.json","graph_json":"https://pith.science/api/pith-number/OVVGBTUGBGUZGHQYVR5TBYUA3Q/graph.json","events_json":"https://pith.science/api/pith-number/OVVGBTUGBGUZGHQYVR5TBYUA3Q/events.json","paper":"https://pith.science/paper/OVVGBTUG"},"agent_actions":{"view_html":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q","download_json":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q.json","view_paper":"https://pith.science/paper/OVVGBTUG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.03021&json=true","fetch_graph":"https://pith.science/api/pith-number/OVVGBTUGBGUZGHQYVR5TBYUA3Q/graph.json","fetch_events":"https://pith.science/api/pith-number/OVVGBTUGBGUZGHQYVR5TBYUA3Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q/action/storage_attestation","attest_author":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q/action/author_attestation","sign_citation":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q/action/citation_signature","submit_replication":"https://pith.science/pith/OVVGBTUGBGUZGHQYVR5TBYUA3Q/action/replication_record"}},"created_at":"2026-08-05T00:44:37.738103+00:00","updated_at":"2026-08-05T00:44:37.738103+00:00"}