{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MFHSSBTWXMC2RKAQLRGIMJWBUH","short_pith_number":"pith:MFHSSBTW","schema_version":"1.0","canonical_sha256":"614f290676bb05a8a8105c4c8626c1a1c3b1c3ac5c59b856c81deaaac60d0750","source":{"kind":"arxiv","id":"2501.17790","version":1},"attestation_state":"computed","paper":{"title":"BreezyVoice: Adapting TTS for Taiwanese Mandarin with Enhanced Polyphone Disambiguation -- Challenges and Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chan-Jan Hsu, Chen-An Li, Chia-Chun Lin, Chien-Cheng Chen, Chien-Yu Yu, Da-shan Shiu, Ho Lam Chung, Hung-yi Lee, Ming-Ji Lee, Ru-Heng Huang, Wei-Chih Chen, Yi-Chang Chen, Yi-Cheng Lin","submitted_at":"2025-01-29T17:31:26Z","abstract_excerpt":"We present BreezyVoice, a Text-to-Speech (TTS) system specifically adapted for Taiwanese Mandarin, highlighting phonetic control abilities to address the unique challenges of polyphone disambiguation in the language. Building upon CosyVoice, we incorporate a $S^{3}$ tokenizer, a large language model (LLM), an optimal-transport conditional flow matching model (OT-CFM), and a grapheme to phoneme prediction model, to generate realistic speech that closely mimics human utterances. Our evaluation demonstrates BreezyVoice's superior performance in both general and code-switching contexts, highlighti"},"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":"2501.17790","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-29T17:31:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6020f8e46486dc27b3c2bc26d38e98d4eb01a3d4283ebffd1cdf332073b60499","abstract_canon_sha256":"7e70bf3aab13aff2b46151b945f32fbae80c1155f841a43c22338bff3b3f90ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:57.904816Z","signature_b64":"3uV66UlVRdTksqODrwGN76qB6m9dhDL6sh/10QD+Aip/KgtrqgWBwHhbLSQfLn0Mc3fS3yAkqhfWKn+Zh3NvCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"614f290676bb05a8a8105c4c8626c1a1c3b1c3ac5c59b856c81deaaac60d0750","last_reissued_at":"2026-07-05T10:06:57.904363Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:57.904363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BreezyVoice: Adapting TTS for Taiwanese Mandarin with Enhanced Polyphone Disambiguation -- Challenges and Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chan-Jan Hsu, Chen-An Li, Chia-Chun Lin, Chien-Cheng Chen, Chien-Yu Yu, Da-shan Shiu, Ho Lam Chung, Hung-yi Lee, Ming-Ji Lee, Ru-Heng Huang, Wei-Chih Chen, Yi-Chang Chen, Yi-Cheng Lin","submitted_at":"2025-01-29T17:31:26Z","abstract_excerpt":"We present BreezyVoice, a Text-to-Speech (TTS) system specifically adapted for Taiwanese Mandarin, highlighting phonetic control abilities to address the unique challenges of polyphone disambiguation in the language. Building upon CosyVoice, we incorporate a $S^{3}$ tokenizer, a large language model (LLM), an optimal-transport conditional flow matching model (OT-CFM), and a grapheme to phoneme prediction model, to generate realistic speech that closely mimics human utterances. Our evaluation demonstrates BreezyVoice's superior performance in both general and code-switching contexts, highlighti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17790","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/2501.17790/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":"2501.17790","created_at":"2026-07-05T10:06:57.904420+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.17790v1","created_at":"2026-07-05T10:06:57.904420+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17790","created_at":"2026-07-05T10:06:57.904420+00:00"},{"alias_kind":"pith_short_12","alias_value":"MFHSSBTWXMC2","created_at":"2026-07-05T10:06:57.904420+00:00"},{"alias_kind":"pith_short_16","alias_value":"MFHSSBTWXMC2RKAQ","created_at":"2026-07-05T10:06:57.904420+00:00"},{"alias_kind":"pith_short_8","alias_value":"MFHSSBTW","created_at":"2026-07-05T10:06:57.904420+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21215","citing_title":"Speaker Identity in Non-Verbal Vocalizations: Conditional Distillation and Mixture of Experts Approach","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26347","citing_title":"The False Resonance: A Critical Examination of Emotion Embedding Similarity for Speech Generation Evaluation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10065","citing_title":"ASPIRin: Action Space Projection for Interactivity-Optimized Reinforcement Learning in Full-Duplex Speech Language Models","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH","json":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH.json","graph_json":"https://pith.science/api/pith-number/MFHSSBTWXMC2RKAQLRGIMJWBUH/graph.json","events_json":"https://pith.science/api/pith-number/MFHSSBTWXMC2RKAQLRGIMJWBUH/events.json","paper":"https://pith.science/paper/MFHSSBTW"},"agent_actions":{"view_html":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH","download_json":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH.json","view_paper":"https://pith.science/paper/MFHSSBTW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.17790&json=true","fetch_graph":"https://pith.science/api/pith-number/MFHSSBTWXMC2RKAQLRGIMJWBUH/graph.json","fetch_events":"https://pith.science/api/pith-number/MFHSSBTWXMC2RKAQLRGIMJWBUH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH/action/storage_attestation","attest_author":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH/action/author_attestation","sign_citation":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH/action/citation_signature","submit_replication":"https://pith.science/pith/MFHSSBTWXMC2RKAQLRGIMJWBUH/action/replication_record"}},"created_at":"2026-07-05T10:06:57.904420+00:00","updated_at":"2026-07-05T10:06:57.904420+00:00"}