{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:5IEJFG6K2H4GTXXS3HFJV4JI4C","short_pith_number":"pith:5IEJFG6K","schema_version":"1.0","canonical_sha256":"ea08929bcad1f869def2d9ca9af128e08216b7efb4b76b975f62f7f04b3aa579","source":{"kind":"arxiv","id":"1809.08895","version":3},"attestation_state":"computed","paper":{"title":"Neural Speech Synthesis with Transformer Network","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ming Liu, Ming Zhou, Naihan Li, Sheng Zhao, Shujie Liu, Yanqing Liu","submitted_at":"2018-09-19T07:41:17Z","abstract_excerpt":"Although end-to-end neural text-to-speech (TTS) methods (such as Tacotron2) are proposed and achieve state-of-the-art performance, they still suffer from two problems: 1) low efficiency during training and inference; 2) hard to model long dependency using current recurrent neural networks (RNNs). Inspired by the success of Transformer network in neural machine translation (NMT), in this paper, we introduce and adapt the multi-head attention mechanism to replace the RNN structures and also the original attention mechanism in Tacotron2. With the help of multi-head self-attention, the hidden stat"},"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":"1809.08895","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-09-19T07:41:17Z","cross_cats_sorted":[],"title_canon_sha256":"0a6370f3db3458bc08cd381eea478da221cd22add8e4304572b56409bd860c13","abstract_canon_sha256":"a48323fe6709a4431229a1bee30f2658339efa51387ce836cef8b38ee73fb004"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:07.377297Z","signature_b64":"eze1pDwo2n9detGs5VOJ/ufRy04YstyJj0NdpRaCbyZ2PEq8FgpyGxhenXqKG+aiFVC1DFSB/6wkxHn6C4LPAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea08929bcad1f869def2d9ca9af128e08216b7efb4b76b975f62f7f04b3aa579","last_reissued_at":"2026-05-17T23:55:07.376841Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:07.376841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Speech Synthesis with Transformer Network","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ming Liu, Ming Zhou, Naihan Li, Sheng Zhao, Shujie Liu, Yanqing Liu","submitted_at":"2018-09-19T07:41:17Z","abstract_excerpt":"Although end-to-end neural text-to-speech (TTS) methods (such as Tacotron2) are proposed and achieve state-of-the-art performance, they still suffer from two problems: 1) low efficiency during training and inference; 2) hard to model long dependency using current recurrent neural networks (RNNs). Inspired by the success of Transformer network in neural machine translation (NMT), in this paper, we introduce and adapt the multi-head attention mechanism to replace the RNN structures and also the original attention mechanism in Tacotron2. With the help of multi-head self-attention, the hidden stat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.08895","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":""},"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":"1809.08895","created_at":"2026-05-17T23:55:07.376916+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.08895v3","created_at":"2026-05-17T23:55:07.376916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.08895","created_at":"2026-05-17T23:55:07.376916+00:00"},{"alias_kind":"pith_short_12","alias_value":"5IEJFG6K2H4G","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"5IEJFG6K2H4GTXXS","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"5IEJFG6K","created_at":"2026-05-18T12:32:08.215937+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1907.02479","citing_title":"Fine-grained robust prosody transfer for single-speaker neural text-to-speech","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"1907.09006","citing_title":"Forward-Backward Decoding for Regularizing End-to-End TTS","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C","json":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C.json","graph_json":"https://pith.science/api/pith-number/5IEJFG6K2H4GTXXS3HFJV4JI4C/graph.json","events_json":"https://pith.science/api/pith-number/5IEJFG6K2H4GTXXS3HFJV4JI4C/events.json","paper":"https://pith.science/paper/5IEJFG6K"},"agent_actions":{"view_html":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C","download_json":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C.json","view_paper":"https://pith.science/paper/5IEJFG6K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.08895&json=true","fetch_graph":"https://pith.science/api/pith-number/5IEJFG6K2H4GTXXS3HFJV4JI4C/graph.json","fetch_events":"https://pith.science/api/pith-number/5IEJFG6K2H4GTXXS3HFJV4JI4C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C/action/storage_attestation","attest_author":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C/action/author_attestation","sign_citation":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C/action/citation_signature","submit_replication":"https://pith.science/pith/5IEJFG6K2H4GTXXS3HFJV4JI4C/action/replication_record"}},"created_at":"2026-05-17T23:55:07.376916+00:00","updated_at":"2026-05-17T23:55:07.376916+00:00"}