{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:RXW4H4LT22QYPU2O726TS3RP5Q","short_pith_number":"pith:RXW4H4LT","schema_version":"1.0","canonical_sha256":"8dedc3f173d6a187d34efebd396e2fec361019ab37b873badfca8069b88241ac","source":{"kind":"arxiv","id":"1808.10128","version":1},"attestation_state":"computed","paper":{"title":"Semi-Supervised Training for Improving Data Efficiency in End-to-End Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"RJ Skerry-Ryan, Wei-Ning Hsu, Yu-An Chung, Yuxuan Wang, Yu Zhang","submitted_at":"2018-08-30T05:51:30Z","abstract_excerpt":"Although end-to-end text-to-speech (TTS) models such as Tacotron have shown excellent results, they typically require a sizable set of high-quality <text, audio> pairs for training, which are expensive to collect. In this paper, we propose a semi-supervised training framework to improve the data efficiency of Tacotron. The idea is to allow Tacotron to utilize textual and acoustic knowledge contained in large, publicly-available text and speech corpora. Importantly, these external data are unpaired and potentially noisy. Specifically, first we embed each word in the input text into word vectors"},"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":"1808.10128","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-30T05:51:30Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"f1d6a4408faadafcab9174a2c3443d57ba3974b420afd5da4155d63a714588d5","abstract_canon_sha256":"b359785ca930cfcf8cdc0eeb33d55ba5adf2b216fe0593de7b5eda950064091f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:06:49.746623Z","signature_b64":"UkAW9CObMk4N6eia6RIzk2KZzqxADPlDDK3/5jB8R+e95TyUIQmNmYC6yVV7m4T+vngjOmcIdLnis722tb7iCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8dedc3f173d6a187d34efebd396e2fec361019ab37b873badfca8069b88241ac","last_reissued_at":"2026-05-18T00:06:49.745902Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:06:49.745902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Semi-Supervised Training for Improving Data Efficiency in End-to-End Speech Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"RJ Skerry-Ryan, Wei-Ning Hsu, Yu-An Chung, Yuxuan Wang, Yu Zhang","submitted_at":"2018-08-30T05:51:30Z","abstract_excerpt":"Although end-to-end text-to-speech (TTS) models such as Tacotron have shown excellent results, they typically require a sizable set of high-quality <text, audio> pairs for training, which are expensive to collect. In this paper, we propose a semi-supervised training framework to improve the data efficiency of Tacotron. The idea is to allow Tacotron to utilize textual and acoustic knowledge contained in large, publicly-available text and speech corpora. Importantly, these external data are unpaired and potentially noisy. Specifically, first we embed each word in the input text into word vectors"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.10128","kind":"arxiv","version":1},"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":"1808.10128","created_at":"2026-05-18T00:06:49.746016+00:00"},{"alias_kind":"arxiv_version","alias_value":"1808.10128v1","created_at":"2026-05-18T00:06:49.746016+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.10128","created_at":"2026-05-18T00:06:49.746016+00:00"},{"alias_kind":"pith_short_12","alias_value":"RXW4H4LT22QY","created_at":"2026-05-18T12:32:50.500415+00:00"},{"alias_kind":"pith_short_16","alias_value":"RXW4H4LT22QYPU2O","created_at":"2026-05-18T12:32:50.500415+00:00"},{"alias_kind":"pith_short_8","alias_value":"RXW4H4LT","created_at":"2026-05-18T12:32:50.500415+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.09006","citing_title":"Forward-Backward Decoding for Regularizing End-to-End TTS","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q","json":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q.json","graph_json":"https://pith.science/api/pith-number/RXW4H4LT22QYPU2O726TS3RP5Q/graph.json","events_json":"https://pith.science/api/pith-number/RXW4H4LT22QYPU2O726TS3RP5Q/events.json","paper":"https://pith.science/paper/RXW4H4LT"},"agent_actions":{"view_html":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q","download_json":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q.json","view_paper":"https://pith.science/paper/RXW4H4LT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1808.10128&json=true","fetch_graph":"https://pith.science/api/pith-number/RXW4H4LT22QYPU2O726TS3RP5Q/graph.json","fetch_events":"https://pith.science/api/pith-number/RXW4H4LT22QYPU2O726TS3RP5Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q/action/storage_attestation","attest_author":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q/action/author_attestation","sign_citation":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q/action/citation_signature","submit_replication":"https://pith.science/pith/RXW4H4LT22QYPU2O726TS3RP5Q/action/replication_record"}},"created_at":"2026-05-18T00:06:49.746016+00:00","updated_at":"2026-05-18T00:06:49.746016+00:00"}