{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:36QIVER2WUTIJBVJJDYKCKVHDG","short_pith_number":"pith:36QIVER2","schema_version":"1.0","canonical_sha256":"dfa08a923ab5268486a948f0a12aa719a6fbce5f661ed049d007f003e6f698ff","source":{"kind":"arxiv","id":"2607.02633","version":1},"attestation_state":"computed","paper":{"title":"GRAFT: Grafted Reference Audio for Fine-grained Pronunciation in Zero-shot Text-to-Speech","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Antonis Asonitis, Aref Farhadipour, Francesco Verdini, Juan Pablo Zuluaga Gomez, Marzieh Razavi, Pierre-Edouard Honnet, Vijeta Avijeet","submitted_at":"2026-07-02T14:40:21Z","abstract_excerpt":"We present GRAFT, a per-word pronunciation conditioning mechanism for text-to-speech neural codec language modeling. Existing systems reach high intelligibility and naturalness but inherit the ambiguity of text and mispronounce rare proper nouns, loanwords and technical terms. Even phoneme-conditioned models offer no direct acoustic handle for per-word pronunciation. GRAFT controls the pronunciation of a chosen word from a short spoken sample of it, encoded with the model's own speech tokenizer and bound to the word's position in the prompt. Voice conversion during training-data construction d"},"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":"2607.02633","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T14:40:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c6ba479dd4b728559c5b5fb5786ad43abee20b3b812f74e362e6f67f4778803b","abstract_canon_sha256":"582f0ec090fb71f41d69c3fdc720a012cb4262952b4926fd22574008f5ec25b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T00:16:04.529846Z","signature_b64":"4g1RGgtDxWqjhDMA1IxdjPhJiIeKAa/9RpX73U8LF9hpTR07FkU814OsKlQkSLBSXxMqrAdKtgEv6JXX5A8oDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dfa08a923ab5268486a948f0a12aa719a6fbce5f661ed049d007f003e6f698ff","last_reissued_at":"2026-07-07T00:16:04.529249Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T00:16:04.529249Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GRAFT: Grafted Reference Audio for Fine-grained Pronunciation in Zero-shot Text-to-Speech","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Antonis Asonitis, Aref Farhadipour, Francesco Verdini, Juan Pablo Zuluaga Gomez, Marzieh Razavi, Pierre-Edouard Honnet, Vijeta Avijeet","submitted_at":"2026-07-02T14:40:21Z","abstract_excerpt":"We present GRAFT, a per-word pronunciation conditioning mechanism for text-to-speech neural codec language modeling. Existing systems reach high intelligibility and naturalness but inherit the ambiguity of text and mispronounce rare proper nouns, loanwords and technical terms. Even phoneme-conditioned models offer no direct acoustic handle for per-word pronunciation. GRAFT controls the pronunciation of a chosen word from a short spoken sample of it, encoded with the model's own speech tokenizer and bound to the word's position in the prompt. Voice conversion during training-data construction d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02633","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/2607.02633/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":"2607.02633","created_at":"2026-07-07T00:16:04.529350+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02633v1","created_at":"2026-07-07T00:16:04.529350+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02633","created_at":"2026-07-07T00:16:04.529350+00:00"},{"alias_kind":"pith_short_12","alias_value":"36QIVER2WUTI","created_at":"2026-07-07T00:16:04.529350+00:00"},{"alias_kind":"pith_short_16","alias_value":"36QIVER2WUTIJBVJ","created_at":"2026-07-07T00:16:04.529350+00:00"},{"alias_kind":"pith_short_8","alias_value":"36QIVER2","created_at":"2026-07-07T00:16:04.529350+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/36QIVER2WUTIJBVJJDYKCKVHDG","json":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG.json","graph_json":"https://pith.science/api/pith-number/36QIVER2WUTIJBVJJDYKCKVHDG/graph.json","events_json":"https://pith.science/api/pith-number/36QIVER2WUTIJBVJJDYKCKVHDG/events.json","paper":"https://pith.science/paper/36QIVER2"},"agent_actions":{"view_html":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG","download_json":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG.json","view_paper":"https://pith.science/paper/36QIVER2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02633&json=true","fetch_graph":"https://pith.science/api/pith-number/36QIVER2WUTIJBVJJDYKCKVHDG/graph.json","fetch_events":"https://pith.science/api/pith-number/36QIVER2WUTIJBVJJDYKCKVHDG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG/action/storage_attestation","attest_author":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG/action/author_attestation","sign_citation":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG/action/citation_signature","submit_replication":"https://pith.science/pith/36QIVER2WUTIJBVJJDYKCKVHDG/action/replication_record"}},"created_at":"2026-07-07T00:16:04.529350+00:00","updated_at":"2026-07-07T00:16:04.529350+00:00"}