{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2A3XNTRKYW42OBJQOVLBVWT6IV","short_pith_number":"pith:2A3XNTRK","schema_version":"1.0","canonical_sha256":"d03776ce2ac5b9a7053075561ada7e45646b2681913e32ef81a0fa622272c518","source":{"kind":"arxiv","id":"2309.03199","version":2},"attestation_state":"computed","paper":{"title":"Matcha-TTS: A fast TTS architecture with conditional flow matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"\\'Eva Sz\\'ekely, Gustav Eje Henter, Jonas Beskow, Ruibo Tu, Shivam Mehta","submitted_at":"2023-09-06T17:59:57Z","abstract_excerpt":"We introduce Matcha-TTS, a new encoder-decoder architecture for speedy TTS acoustic modelling, trained using optimal-transport conditional flow matching (OT-CFM). This yields an ODE-based decoder capable of high output quality in fewer synthesis steps than models trained using score matching. Careful design choices additionally ensure each synthesis step is fast to run. The method is probabilistic, non-autoregressive, and learns to speak from scratch without external alignments. Compared to strong pre-trained baseline models, the Matcha-TTS system has the smallest memory footprint, rivals the "},"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":"2309.03199","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2023-09-06T17:59:57Z","cross_cats_sorted":["cs.HC","cs.LG","cs.SD"],"title_canon_sha256":"9565457b396ef44b0aae3c2cf99aefd85d32249efe348c25e6109e114c6de9ae","abstract_canon_sha256":"107637c62a5a790554e211e2b92a992be8187bb3e4ffe707f0fe7d42d57e3754"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:31:54.357574Z","signature_b64":"vasWiFJ5OjpgCWTsw/EzWbBROB01qe1zH8QMLWAZOTzmOWY6xiOmIRrQGL5YAaybterlEnfb/KlnNC5kRgQkAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d03776ce2ac5b9a7053075561ada7e45646b2681913e32ef81a0fa622272c518","last_reissued_at":"2026-07-05T07:31:54.357153Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:31:54.357153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Matcha-TTS: A fast TTS architecture with conditional flow matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.HC","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"\\'Eva Sz\\'ekely, Gustav Eje Henter, Jonas Beskow, Ruibo Tu, Shivam Mehta","submitted_at":"2023-09-06T17:59:57Z","abstract_excerpt":"We introduce Matcha-TTS, a new encoder-decoder architecture for speedy TTS acoustic modelling, trained using optimal-transport conditional flow matching (OT-CFM). This yields an ODE-based decoder capable of high output quality in fewer synthesis steps than models trained using score matching. Careful design choices additionally ensure each synthesis step is fast to run. The method is probabilistic, non-autoregressive, and learns to speak from scratch without external alignments. Compared to strong pre-trained baseline models, the Matcha-TTS system has the smallest memory footprint, rivals the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.03199","kind":"arxiv","version":2},"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/2309.03199/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":"2309.03199","created_at":"2026-07-05T07:31:54.357216+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.03199v2","created_at":"2026-07-05T07:31:54.357216+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.03199","created_at":"2026-07-05T07:31:54.357216+00:00"},{"alias_kind":"pith_short_12","alias_value":"2A3XNTRKYW42","created_at":"2026-07-05T07:31:54.357216+00:00"},{"alias_kind":"pith_short_16","alias_value":"2A3XNTRKYW42OBJQ","created_at":"2026-07-05T07:31:54.357216+00:00"},{"alias_kind":"pith_short_8","alias_value":"2A3XNTRK","created_at":"2026-07-05T07:31:54.357216+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.07462","citing_title":"SemBridge: Semantic Token Anchoring for Continuous-Latent Autoregressive Speech Generation","ref_index":96,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV","json":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV.json","graph_json":"https://pith.science/api/pith-number/2A3XNTRKYW42OBJQOVLBVWT6IV/graph.json","events_json":"https://pith.science/api/pith-number/2A3XNTRKYW42OBJQOVLBVWT6IV/events.json","paper":"https://pith.science/paper/2A3XNTRK"},"agent_actions":{"view_html":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV","download_json":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV.json","view_paper":"https://pith.science/paper/2A3XNTRK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.03199&json=true","fetch_graph":"https://pith.science/api/pith-number/2A3XNTRKYW42OBJQOVLBVWT6IV/graph.json","fetch_events":"https://pith.science/api/pith-number/2A3XNTRKYW42OBJQOVLBVWT6IV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV/action/storage_attestation","attest_author":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV/action/author_attestation","sign_citation":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV/action/citation_signature","submit_replication":"https://pith.science/pith/2A3XNTRKYW42OBJQOVLBVWT6IV/action/replication_record"}},"created_at":"2026-07-05T07:31:54.357216+00:00","updated_at":"2026-07-05T07:31:54.357216+00:00"}