{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:P7OR2SBN2BN4H365YIA3M2HGI3","short_pith_number":"pith:P7OR2SBN","schema_version":"1.0","canonical_sha256":"7fdd1d482dd05bc3efddc201b668e646de91ce7588eeacbea14c7a8dd6244f48","source":{"kind":"arxiv","id":"2404.08857","version":2},"attestation_state":"computed","paper":{"title":"Voice Attribute Editing with Text Prompt","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Jia Pan, Li-Juan Liu, Yang Ai, Zhengyan Sheng, Zhen-Hua Ling","submitted_at":"2024-04-13T00:07:40Z","abstract_excerpt":"Despite recent advancements in speech generation with text prompt providing control over speech style, voice attributes in synthesized speech remain elusive and challenging to control. This paper introduces a novel task: voice attribute editing with text prompt, with the goal of making relative modifications to voice attributes according to the actions described in the text prompt. To solve this task, VoxEditor, an end-to-end generative model, is proposed. In VoxEditor, addressing the insufficiency of text prompt, a Residual Memory (ResMem) block is designed, that efficiently maps voice attrib"},"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":"2404.08857","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-04-13T00:07:40Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"7313bc253320917015cf2b53a98ecef784725a2401b83fcf727f73b6dd2ce5b2","abstract_canon_sha256":"324ff834482d2f19c7515a7e2dbc67189d4bc9632ef06dd984e40072fd65d000"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:24.580862Z","signature_b64":"3HHy/B+E2oehsOtaDtDx0N5QOE5ei4pwNEVMgVHNIdlOWNRMOXZ3NKQ1bACE9zkL+jU0cKM+O3k1yRTFajJDDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fdd1d482dd05bc3efddc201b668e646de91ce7588eeacbea14c7a8dd6244f48","last_reissued_at":"2026-07-05T09:42:24.580277Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:24.580277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Voice Attribute Editing with Text Prompt","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Jia Pan, Li-Juan Liu, Yang Ai, Zhengyan Sheng, Zhen-Hua Ling","submitted_at":"2024-04-13T00:07:40Z","abstract_excerpt":"Despite recent advancements in speech generation with text prompt providing control over speech style, voice attributes in synthesized speech remain elusive and challenging to control. This paper introduces a novel task: voice attribute editing with text prompt, with the goal of making relative modifications to voice attributes according to the actions described in the text prompt. To solve this task, VoxEditor, an end-to-end generative model, is proposed. In VoxEditor, addressing the insufficiency of text prompt, a Residual Memory (ResMem) block is designed, that efficiently maps voice attrib"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.08857","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/2404.08857/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":"2404.08857","created_at":"2026-07-05T09:42:24.580367+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.08857v2","created_at":"2026-07-05T09:42:24.580367+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.08857","created_at":"2026-07-05T09:42:24.580367+00:00"},{"alias_kind":"pith_short_12","alias_value":"P7OR2SBN2BN4","created_at":"2026-07-05T09:42:24.580367+00:00"},{"alias_kind":"pith_short_16","alias_value":"P7OR2SBN2BN4H365","created_at":"2026-07-05T09:42:24.580367+00:00"},{"alias_kind":"pith_short_8","alias_value":"P7OR2SBN","created_at":"2026-07-05T09:42:24.580367+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.09382","citing_title":"The Voice Timbre Attribute Detection 2025 Challenge Evaluation Plan","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3","json":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3.json","graph_json":"https://pith.science/api/pith-number/P7OR2SBN2BN4H365YIA3M2HGI3/graph.json","events_json":"https://pith.science/api/pith-number/P7OR2SBN2BN4H365YIA3M2HGI3/events.json","paper":"https://pith.science/paper/P7OR2SBN"},"agent_actions":{"view_html":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3","download_json":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3.json","view_paper":"https://pith.science/paper/P7OR2SBN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.08857&json=true","fetch_graph":"https://pith.science/api/pith-number/P7OR2SBN2BN4H365YIA3M2HGI3/graph.json","fetch_events":"https://pith.science/api/pith-number/P7OR2SBN2BN4H365YIA3M2HGI3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3/action/storage_attestation","attest_author":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3/action/author_attestation","sign_citation":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3/action/citation_signature","submit_replication":"https://pith.science/pith/P7OR2SBN2BN4H365YIA3M2HGI3/action/replication_record"}},"created_at":"2026-07-05T09:42:24.580367+00:00","updated_at":"2026-07-05T09:42:24.580367+00:00"}