{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:X3JCSHWARKVLBFXFKD7TXNX3TF","short_pith_number":"pith:X3JCSHWA","schema_version":"1.0","canonical_sha256":"bed2291ec08aaab096e550ff3bb6fb9956eb12d53680791a253d3910b4ff1947","source":{"kind":"arxiv","id":"2106.12896","version":2},"attestation_state":"computed","paper":{"title":"Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.SD","authors_text":"Abdelhamid Ezzerg, Bartosz Putrycz, Daniel Korzekwa, Goeric Huybrechts, Kamil Pokora, Raahil Shah, Thomas Merritt, Viacheslav Klimkov","submitted_at":"2021-06-24T10:52:10Z","abstract_excerpt":"Whilst recent neural text-to-speech (TTS) approaches produce high-quality speech, they typically require a large amount of recordings from the target speaker. In previous work, a 3-step method was proposed to generate high-quality TTS while greatly reducing the amount of data required for training. However, we have observed a ceiling effect in the level of naturalness achievable for highly expressive voices when using this approach. In this paper, we present a method for building highly expressive TTS voices with as little as 15 minutes of speech data from the target speaker. Compared to the c"},"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":"2106.12896","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2021-06-24T10:52:10Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d01f5d350972228758a5e452500e81bac624dff88e9d459d319974d8a13ba060","abstract_canon_sha256":"3478069c9cbca6872a32d6bebd30bf98868929880a585d6fe2e81c9757b8ac14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:52:28.852847Z","signature_b64":"UR0CKO0l838bT0xiN6Zi0MBVMu8uvH9n751Ox/6AW5uVJQUCCMIwgbTG4gjC2hN4xqC8oYy6omjF7JdkudwDDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bed2291ec08aaab096e550ff3bb6fb9956eb12d53680791a253d3910b4ff1947","last_reissued_at":"2026-07-05T02:52:28.852408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:52:28.852408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.SD","authors_text":"Abdelhamid Ezzerg, Bartosz Putrycz, Daniel Korzekwa, Goeric Huybrechts, Kamil Pokora, Raahil Shah, Thomas Merritt, Viacheslav Klimkov","submitted_at":"2021-06-24T10:52:10Z","abstract_excerpt":"Whilst recent neural text-to-speech (TTS) approaches produce high-quality speech, they typically require a large amount of recordings from the target speaker. In previous work, a 3-step method was proposed to generate high-quality TTS while greatly reducing the amount of data required for training. However, we have observed a ceiling effect in the level of naturalness achievable for highly expressive voices when using this approach. In this paper, we present a method for building highly expressive TTS voices with as little as 15 minutes of speech data from the target speaker. Compared to the c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.12896","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/2106.12896/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":"2106.12896","created_at":"2026-07-05T02:52:28.852478+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.12896v2","created_at":"2026-07-05T02:52:28.852478+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.12896","created_at":"2026-07-05T02:52:28.852478+00:00"},{"alias_kind":"pith_short_12","alias_value":"X3JCSHWARKVL","created_at":"2026-07-05T02:52:28.852478+00:00"},{"alias_kind":"pith_short_16","alias_value":"X3JCSHWARKVLBFXF","created_at":"2026-07-05T02:52:28.852478+00:00"},{"alias_kind":"pith_short_8","alias_value":"X3JCSHWA","created_at":"2026-07-05T02:52:28.852478+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02166","citing_title":"Dhvani: A Weakly-supervised Phonemic Error Detection and Personalized Feedback System for Hindi","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF","json":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF.json","graph_json":"https://pith.science/api/pith-number/X3JCSHWARKVLBFXFKD7TXNX3TF/graph.json","events_json":"https://pith.science/api/pith-number/X3JCSHWARKVLBFXFKD7TXNX3TF/events.json","paper":"https://pith.science/paper/X3JCSHWA"},"agent_actions":{"view_html":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF","download_json":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF.json","view_paper":"https://pith.science/paper/X3JCSHWA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.12896&json=true","fetch_graph":"https://pith.science/api/pith-number/X3JCSHWARKVLBFXFKD7TXNX3TF/graph.json","fetch_events":"https://pith.science/api/pith-number/X3JCSHWARKVLBFXFKD7TXNX3TF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF/action/storage_attestation","attest_author":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF/action/author_attestation","sign_citation":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF/action/citation_signature","submit_replication":"https://pith.science/pith/X3JCSHWARKVLBFXFKD7TXNX3TF/action/replication_record"}},"created_at":"2026-07-05T02:52:28.852478+00:00","updated_at":"2026-07-05T02:52:28.852478+00:00"}