{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:X7UGLHN3DFBUBPWRCTN3C7OSCJ","short_pith_number":"pith:X7UGLHN3","schema_version":"1.0","canonical_sha256":"bfe8659dbb194340bed114dbb17dd21248b5bc2f9a93513e0081af093b0d3f2f","source":{"kind":"arxiv","id":"2508.12001","version":3},"attestation_state":"computed","paper":{"title":"FNH-TTS: Mixture-of-Experts Duration Modeling for Robust Neural Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Huizhi Liang, Limei Yu, Qingliang Meng, Tian Li, Wei Liang, Yuqing Deng","submitted_at":"2025-08-16T10:04:21Z","abstract_excerpt":"Current non-autoregressive (NAR) text-to-speech (TTS) systems still struggle to model diverse and speaker-dependent duration variation. We further observe that richer duration variation can increase the synthesis difficulty of existing HiFi-GAN-based vocoders, leading to spectral artifacts and unstable time-frequency structures. To address these issues, we propose FNH-TTS, a VITS-based end-to-end TTS system with Mixture-of-Experts duration modeling and robust vocoder-side synthesis. Specifically, we introduce a Mixture-of-Experts Duration Predictor (MoE-DP) to capture diverse phoneme duration "},"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":"2508.12001","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2025-08-16T10:04:21Z","cross_cats_sorted":[],"title_canon_sha256":"f9cf378e9dea487149534088e1a33b72daec0c98338817ba4d7c26eac54bb93d","abstract_canon_sha256":"da1a02744c03ca8cfb2f492eb2f1325218f9e95c7daaf6b98435f141f2b39a41"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:05:35.898389Z","signature_b64":"31o0yhacDKlqsW7DfHXCq/J+UDX8XD+keBr9zGTRcPJSjGcEmT4ZklWWVHALfrtFpeOoSuDAJMXbUU18x3VADw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bfe8659dbb194340bed114dbb17dd21248b5bc2f9a93513e0081af093b0d3f2f","last_reissued_at":"2026-05-29T02:05:35.897894Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:05:35.897894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FNH-TTS: Mixture-of-Experts Duration Modeling for Robust Neural Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Huizhi Liang, Limei Yu, Qingliang Meng, Tian Li, Wei Liang, Yuqing Deng","submitted_at":"2025-08-16T10:04:21Z","abstract_excerpt":"Current non-autoregressive (NAR) text-to-speech (TTS) systems still struggle to model diverse and speaker-dependent duration variation. We further observe that richer duration variation can increase the synthesis difficulty of existing HiFi-GAN-based vocoders, leading to spectral artifacts and unstable time-frequency structures. To address these issues, we propose FNH-TTS, a VITS-based end-to-end TTS system with Mixture-of-Experts duration modeling and robust vocoder-side synthesis. Specifically, we introduce a Mixture-of-Experts Duration Predictor (MoE-DP) to capture diverse phoneme duration "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.12001","kind":"arxiv","version":3},"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/2508.12001/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":"2508.12001","created_at":"2026-05-29T02:05:35.897953+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.12001v3","created_at":"2026-05-29T02:05:35.897953+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.12001","created_at":"2026-05-29T02:05:35.897953+00:00"},{"alias_kind":"pith_short_12","alias_value":"X7UGLHN3DFBU","created_at":"2026-05-29T02:05:35.897953+00:00"},{"alias_kind":"pith_short_16","alias_value":"X7UGLHN3DFBUBPWR","created_at":"2026-05-29T02:05:35.897953+00:00"},{"alias_kind":"pith_short_8","alias_value":"X7UGLHN3","created_at":"2026-05-29T02:05:35.897953+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/X7UGLHN3DFBUBPWRCTN3C7OSCJ","json":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ.json","graph_json":"https://pith.science/api/pith-number/X7UGLHN3DFBUBPWRCTN3C7OSCJ/graph.json","events_json":"https://pith.science/api/pith-number/X7UGLHN3DFBUBPWRCTN3C7OSCJ/events.json","paper":"https://pith.science/paper/X7UGLHN3"},"agent_actions":{"view_html":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ","download_json":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ.json","view_paper":"https://pith.science/paper/X7UGLHN3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.12001&json=true","fetch_graph":"https://pith.science/api/pith-number/X7UGLHN3DFBUBPWRCTN3C7OSCJ/graph.json","fetch_events":"https://pith.science/api/pith-number/X7UGLHN3DFBUBPWRCTN3C7OSCJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ/action/storage_attestation","attest_author":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ/action/author_attestation","sign_citation":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ/action/citation_signature","submit_replication":"https://pith.science/pith/X7UGLHN3DFBUBPWRCTN3C7OSCJ/action/replication_record"}},"created_at":"2026-05-29T02:05:35.897953+00:00","updated_at":"2026-05-29T02:05:35.897953+00:00"}