{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:ZYVHMQVUV7KTESDBFIDD4E57P3","short_pith_number":"pith:ZYVHMQVU","schema_version":"1.0","canonical_sha256":"ce2a7642b4afd53248612a063e13bf7eef9688432d8dd3ff8236ade092d48858","source":{"kind":"arxiv","id":"1712.05884","version":2},"attestation_state":"computed","paper":{"title":"Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jonathan Shen, Mike Schuster, Navdeep Jaitly, Rif A. Saurous, RJ Skerry-Ryan, Ron J. Weiss, Ruoming Pang, Yannis Agiomyrgiannakis, Yonghui Wu, Yuxuan Wang, Yu Zhang, Zhifeng Chen, Zongheng Yang","submitted_at":"2017-12-16T00:51:40Z","abstract_excerpt":"This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinion score (MOS) of $4.53$ comparable to a MOS of $4.58$ for professionally recorded speech. To validate our design choices, we present ablation studies of key components of our system and evaluate the impa"},"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":"1712.05884","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-12-16T00:51:40Z","cross_cats_sorted":[],"title_canon_sha256":"2175efb1d30bd526346bb59e8974b510aa377e5588bb8b47c6d8b16faac6aee0","abstract_canon_sha256":"d700e588ec6b008bd23fc1571338214b35438914b62b6b11a4c3e4aae70c3918"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:23:13.650372Z","signature_b64":"M/LVYxsDsz4xxDsaWrdZngVmBAUsyAoNU3w+D/qHLLBXzMoR9lwa9WFMo3Pkeil/BLHRpVUOnz0mAOIiMnpRCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce2a7642b4afd53248612a063e13bf7eef9688432d8dd3ff8236ade092d48858","last_reissued_at":"2026-05-18T00:23:13.649719Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:23:13.649719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jonathan Shen, Mike Schuster, Navdeep Jaitly, Rif A. Saurous, RJ Skerry-Ryan, Ron J. Weiss, Ruoming Pang, Yannis Agiomyrgiannakis, Yonghui Wu, Yuxuan Wang, Yu Zhang, Zhifeng Chen, Zongheng Yang","submitted_at":"2017-12-16T00:51:40Z","abstract_excerpt":"This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinion score (MOS) of $4.53$ comparable to a MOS of $4.58$ for professionally recorded speech. To validate our design choices, we present ablation studies of key components of our system and evaluate the impa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.05884","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":""},"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":"1712.05884","created_at":"2026-05-18T00:23:13.649819+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.05884v2","created_at":"2026-05-18T00:23:13.649819+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.05884","created_at":"2026-05-18T00:23:13.649819+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYVHMQVUV7KT","created_at":"2026-05-18T12:31:59.375834+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYVHMQVUV7KTESDB","created_at":"2026-05-18T12:31:59.375834+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYVHMQVU","created_at":"2026-05-18T12:31:59.375834+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1907.07769","citing_title":"Hierarchical Sequence to Sequence Voice Conversion with Limited Data","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20299","citing_title":"Mechanisms of Misgeneralization in Physical Sequence Modeling","ref_index":135,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3","json":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3.json","graph_json":"https://pith.science/api/pith-number/ZYVHMQVUV7KTESDBFIDD4E57P3/graph.json","events_json":"https://pith.science/api/pith-number/ZYVHMQVUV7KTESDBFIDD4E57P3/events.json","paper":"https://pith.science/paper/ZYVHMQVU"},"agent_actions":{"view_html":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3","download_json":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3.json","view_paper":"https://pith.science/paper/ZYVHMQVU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.05884&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYVHMQVUV7KTESDBFIDD4E57P3/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYVHMQVUV7KTESDBFIDD4E57P3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3/action/storage_attestation","attest_author":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3/action/author_attestation","sign_citation":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3/action/citation_signature","submit_replication":"https://pith.science/pith/ZYVHMQVUV7KTESDBFIDD4E57P3/action/replication_record"}},"created_at":"2026-05-18T00:23:13.649819+00:00","updated_at":"2026-05-18T00:23:13.649819+00:00"}