{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4V6DBGFCHD3QQZMSE2KXJN5ID3","short_pith_number":"pith:4V6DBGFC","schema_version":"1.0","canonical_sha256":"e57c3098a238f7086592269574b7a81ee9a6128eb4171ec72765e7e639d44c1c","source":{"kind":"arxiv","id":"2005.05957","version":3},"attestation_state":"computed","paper":{"title":"Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Bryan Catanzaro, Kevin Shih, Rafael Valle, Ryan Prenger","submitted_at":"2020-05-12T17:57:17Z","abstract_excerpt":"In this paper we propose Flowtron: an autoregressive flow-based generative network for text-to-speech synthesis with control over speech variation and style transfer. Flowtron borrows insights from IAF and revamps Tacotron in order to provide high-quality and expressive mel-spectrogram synthesis. Flowtron is optimized by maximizing the likelihood of the training data, which makes training simple and stable. Flowtron learns an invertible mapping of data to a latent space that can be manipulated to control many aspects of speech synthesis (pitch, tone, speech rate, cadence, accent). Our mean opi"},"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":"2005.05957","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2020-05-12T17:57:17Z","cross_cats_sorted":["cs.CL","cs.LG","eess.AS"],"title_canon_sha256":"215b26c17e022b2c648b79faeaca9291f0ad29ac6d555c0a285a342ab333cea0","abstract_canon_sha256":"775ab618869a56f732a7a14e608d777d25b768a9042ea00a9dde0054cc08ad0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:19:39.783334Z","signature_b64":"W8ALzl/D9BI9S7qOSjIJReMxKsvqb1PoagFDRVv/k+1Ikcu61ZUjE8Rel6xdqqs2KK/1I/j9NevnuVhXA/bxCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e57c3098a238f7086592269574b7a81ee9a6128eb4171ec72765e7e639d44c1c","last_reissued_at":"2026-07-05T01:19:39.782844Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:19:39.782844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Bryan Catanzaro, Kevin Shih, Rafael Valle, Ryan Prenger","submitted_at":"2020-05-12T17:57:17Z","abstract_excerpt":"In this paper we propose Flowtron: an autoregressive flow-based generative network for text-to-speech synthesis with control over speech variation and style transfer. Flowtron borrows insights from IAF and revamps Tacotron in order to provide high-quality and expressive mel-spectrogram synthesis. Flowtron is optimized by maximizing the likelihood of the training data, which makes training simple and stable. Flowtron learns an invertible mapping of data to a latent space that can be manipulated to control many aspects of speech synthesis (pitch, tone, speech rate, cadence, accent). Our mean opi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.05957","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/2005.05957/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":"2005.05957","created_at":"2026-07-05T01:19:39.782901+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.05957v3","created_at":"2026-07-05T01:19:39.782901+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.05957","created_at":"2026-07-05T01:19:39.782901+00:00"},{"alias_kind":"pith_short_12","alias_value":"4V6DBGFCHD3Q","created_at":"2026-07-05T01:19:39.782901+00:00"},{"alias_kind":"pith_short_16","alias_value":"4V6DBGFCHD3QQZMS","created_at":"2026-07-05T01:19:39.782901+00:00"},{"alias_kind":"pith_short_8","alias_value":"4V6DBGFC","created_at":"2026-07-05T01:19:39.782901+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05196","citing_title":"Unified Audio Intelligence Without Regressing on Text Intelligence","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2009.09761","citing_title":"DiffWave: A Versatile Diffusion Model for Audio Synthesis","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3","json":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3.json","graph_json":"https://pith.science/api/pith-number/4V6DBGFCHD3QQZMSE2KXJN5ID3/graph.json","events_json":"https://pith.science/api/pith-number/4V6DBGFCHD3QQZMSE2KXJN5ID3/events.json","paper":"https://pith.science/paper/4V6DBGFC"},"agent_actions":{"view_html":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3","download_json":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3.json","view_paper":"https://pith.science/paper/4V6DBGFC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.05957&json=true","fetch_graph":"https://pith.science/api/pith-number/4V6DBGFCHD3QQZMSE2KXJN5ID3/graph.json","fetch_events":"https://pith.science/api/pith-number/4V6DBGFCHD3QQZMSE2KXJN5ID3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3/action/storage_attestation","attest_author":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3/action/author_attestation","sign_citation":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3/action/citation_signature","submit_replication":"https://pith.science/pith/4V6DBGFCHD3QQZMSE2KXJN5ID3/action/replication_record"}},"created_at":"2026-07-05T01:19:39.782901+00:00","updated_at":"2026-07-05T01:19:39.782901+00:00"}