{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YLLEDNKZ3ZO5FBKFNFVCG2PRJK","short_pith_number":"pith:YLLEDNKZ","schema_version":"1.0","canonical_sha256":"c2d641b559de5dd28545696a2369f14ab24b996bf4bd5c86f491e6c3a43bf0a9","source":{"kind":"arxiv","id":"2608.11650","version":1},"attestation_state":"computed","paper":{"title":"Confucius4-TTS: Transcript-Free Cross-Lingual Zero-Shot TTS with a Learnable Speaker Encoder","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Huaxuan Wang, Huimin Wang, Ruiyu Zhang, Yingjie Li, Yitao Duan","submitted_at":"2026-08-12T04:48:29Z","abstract_excerpt":"Recent advances in zero-shot text-to-speech (TTS) have substantially improved speech quality and voice cloning fidelity. However, many zero-shot TTS systems still depend on audio prompt transcripts at inference time. This dependency limits cross-lingual voice cloning, since in-the-wild reference audio is often untranscribed. In this technical report, we present Confucius4-TTS, a multilingual zero-shot TTS system that supports 14 languages and performs both intra-lingual and cross-lingual reference cloning without requiring transcripts of audio prompts. Confucius4-TTS follows a two-stage archit"},"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":"2608.11650","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-08-12T04:48:29Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9a24b0abaae47a59f584192d5ab595a6fc35bcaf688bc3526ab147567eb14537","abstract_canon_sha256":"89eb722b47af528139a02642260c15c8a3d11a307e87e8bdb14a4de981fe098d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T01:27:13.172291Z","signature_b64":"toeYuDbN/cnKW9VtAdo/WfTNhfrOUOjSNWg1rZM5QttJL7/l8PwqF9ScUCDGINT0/UvjV658FyIWcJsSd2SBBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2d641b559de5dd28545696a2369f14ab24b996bf4bd5c86f491e6c3a43bf0a9","last_reissued_at":"2026-08-13T01:27:13.169647Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T01:27:13.169647Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Confucius4-TTS: Transcript-Free Cross-Lingual Zero-Shot TTS with a Learnable Speaker Encoder","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Huaxuan Wang, Huimin Wang, Ruiyu Zhang, Yingjie Li, Yitao Duan","submitted_at":"2026-08-12T04:48:29Z","abstract_excerpt":"Recent advances in zero-shot text-to-speech (TTS) have substantially improved speech quality and voice cloning fidelity. However, many zero-shot TTS systems still depend on audio prompt transcripts at inference time. This dependency limits cross-lingual voice cloning, since in-the-wild reference audio is often untranscribed. In this technical report, we present Confucius4-TTS, a multilingual zero-shot TTS system that supports 14 languages and performs both intra-lingual and cross-lingual reference cloning without requiring transcripts of audio prompts. Confucius4-TTS follows a two-stage archit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11650","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/2608.11650/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":"2608.11650","created_at":"2026-08-13T01:27:13.170728+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.11650v1","created_at":"2026-08-13T01:27:13.170728+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11650","created_at":"2026-08-13T01:27:13.170728+00:00"},{"alias_kind":"pith_short_12","alias_value":"YLLEDNKZ3ZO5","created_at":"2026-08-13T01:27:13.170728+00:00"},{"alias_kind":"pith_short_16","alias_value":"YLLEDNKZ3ZO5FBKF","created_at":"2026-08-13T01:27:13.170728+00:00"},{"alias_kind":"pith_short_8","alias_value":"YLLEDNKZ","created_at":"2026-08-13T01:27:13.170728+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/YLLEDNKZ3ZO5FBKFNFVCG2PRJK","json":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK.json","graph_json":"https://pith.science/api/pith-number/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/graph.json","events_json":"https://pith.science/api/pith-number/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/events.json","paper":"https://pith.science/paper/YLLEDNKZ"},"agent_actions":{"view_html":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK","download_json":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK.json","view_paper":"https://pith.science/paper/YLLEDNKZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.11650&json=true","fetch_graph":"https://pith.science/api/pith-number/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/graph.json","fetch_events":"https://pith.science/api/pith-number/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/action/storage_attestation","attest_author":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/action/author_attestation","sign_citation":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/action/citation_signature","submit_replication":"https://pith.science/pith/YLLEDNKZ3ZO5FBKFNFVCG2PRJK/action/replication_record"}},"created_at":"2026-08-13T01:27:13.170728+00:00","updated_at":"2026-08-13T01:27:13.170728+00:00"}