{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WBZT7YJKLARN2DZXBSWX3NUYAH","short_pith_number":"pith:WBZT7YJK","schema_version":"1.0","canonical_sha256":"b0733fe12a5822dd0f370cad7db69801c0731ca83b7209d5d402a7a6c6acc6c7","source":{"kind":"arxiv","id":"2606.22009","version":1},"attestation_state":"computed","paper":{"title":"Benchmarking Large Language Models for Grapheme-to-Phoneme Conversion: A Japanese Case Study","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Tomoki Koriyama","submitted_at":"2026-06-20T12:17:03Z","abstract_excerpt":"Grapheme-to-phoneme (G2P) conversion is essential for controllable and robust text-to-speech, and large language models (LLMs), with broad linguistic knowledge, offer a promising approach. We benchmarked over 30 LLMs on Japanese G2P, comparing them with conventional morphological analyzers on 3000 manually annotated sentences. We evaluated two prompting strategies: a parse mode, where the LLM performs morphological analysis followed by rule-based kana conversion, and a direct mode, where the LLM directly predicts kana readings. The results show that model size, version, and Japanese-specialize"},"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":"2606.22009","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-20T12:17:03Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"4e469c8f0c6b3ea1290da391b168b36898c4a036e6eb3bc3a8c97b64d46ab7f3","abstract_canon_sha256":"4b350322981d7790d4d8142f4d2780cde0eb9a715ba4de2f63e268e449eeed80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:05.484797Z","signature_b64":"dbgRyJOZdOTJQ8R3pXVUfpT21/fe6c81R1WiWz4ID9Bw1kPlGDDQCUfbnCY6liB7oeOHvH9joO3Rbzh44cTADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0733fe12a5822dd0f370cad7db69801c0731ca83b7209d5d402a7a6c6acc6c7","last_reissued_at":"2026-06-23T02:13:05.484410Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:05.484410Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Benchmarking Large Language Models for Grapheme-to-Phoneme Conversion: A Japanese Case Study","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Tomoki Koriyama","submitted_at":"2026-06-20T12:17:03Z","abstract_excerpt":"Grapheme-to-phoneme (G2P) conversion is essential for controllable and robust text-to-speech, and large language models (LLMs), with broad linguistic knowledge, offer a promising approach. We benchmarked over 30 LLMs on Japanese G2P, comparing them with conventional morphological analyzers on 3000 manually annotated sentences. We evaluated two prompting strategies: a parse mode, where the LLM performs morphological analysis followed by rule-based kana conversion, and a direct mode, where the LLM directly predicts kana readings. The results show that model size, version, and Japanese-specialize"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22009","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/2606.22009/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":"2606.22009","created_at":"2026-06-23T02:13:05.484467+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22009v1","created_at":"2026-06-23T02:13:05.484467+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22009","created_at":"2026-06-23T02:13:05.484467+00:00"},{"alias_kind":"pith_short_12","alias_value":"WBZT7YJKLARN","created_at":"2026-06-23T02:13:05.484467+00:00"},{"alias_kind":"pith_short_16","alias_value":"WBZT7YJKLARN2DZX","created_at":"2026-06-23T02:13:05.484467+00:00"},{"alias_kind":"pith_short_8","alias_value":"WBZT7YJK","created_at":"2026-06-23T02:13:05.484467+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.22009","citing_title":"Benchmarking Large Language Models for Grapheme-to-Phoneme Conversion: A Japanese Case Study","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH","json":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH.json","graph_json":"https://pith.science/api/pith-number/WBZT7YJKLARN2DZXBSWX3NUYAH/graph.json","events_json":"https://pith.science/api/pith-number/WBZT7YJKLARN2DZXBSWX3NUYAH/events.json","paper":"https://pith.science/paper/WBZT7YJK"},"agent_actions":{"view_html":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH","download_json":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH.json","view_paper":"https://pith.science/paper/WBZT7YJK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22009&json=true","fetch_graph":"https://pith.science/api/pith-number/WBZT7YJKLARN2DZXBSWX3NUYAH/graph.json","fetch_events":"https://pith.science/api/pith-number/WBZT7YJKLARN2DZXBSWX3NUYAH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH/action/storage_attestation","attest_author":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH/action/author_attestation","sign_citation":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH/action/citation_signature","submit_replication":"https://pith.science/pith/WBZT7YJKLARN2DZXBSWX3NUYAH/action/replication_record"}},"created_at":"2026-06-23T02:13:05.484467+00:00","updated_at":"2026-06-23T02:13:05.484467+00:00"}