{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:672EZKVYEADTXCDJECQ66ASFNX","short_pith_number":"pith:672EZKVY","schema_version":"1.0","canonical_sha256":"f7f44caab820073b886920a1ef02456de3cc1a08c96a4a2d9de90d8762b1de50","source":{"kind":"arxiv","id":"2008.00768","version":1},"attestation_state":"computed","paper":{"title":"One Model, Many Languages: Meta-learning for Multilingual Text-to-Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"eess.AS","authors_text":"Ond\\v{r}ej Du\\v{s}ek, Tom\\'a\\v{s} Nekvinda","submitted_at":"2020-08-03T10:43:30Z","abstract_excerpt":"We introduce an approach to multilingual speech synthesis which uses the meta-learning concept of contextual parameter generation and produces natural-sounding multilingual speech using more languages and less training data than previous approaches. Our model is based on Tacotron 2 with a fully convolutional input text encoder whose weights are predicted by a separate parameter generator network. To boost voice cloning, the model uses an adversarial speaker classifier with a gradient reversal layer that removes speaker-specific information from the encoder.\n  We arranged two experiments to com"},"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":"2008.00768","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-08-03T10:43:30Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"04e847cee0447913593abf30f1e1f139364a509f0b935cd2e81d0ff79e986f99","abstract_canon_sha256":"7cc9872649ec1330ddad376ec051deced6884410f73ccedbbef544b22196569c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:24:10.894178Z","signature_b64":"/Rb3pX9bDXWIhWrYEjcMFyZK8YeN950PgNM2zwlFEsgtJQE1qAwEztkzpcR9O+BsbBORBjFiACssRkopeq9ZAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7f44caab820073b886920a1ef02456de3cc1a08c96a4a2d9de90d8762b1de50","last_reissued_at":"2026-07-05T01:24:10.893667Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:24:10.893667Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"One Model, Many Languages: Meta-learning for Multilingual Text-to-Speech","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"eess.AS","authors_text":"Ond\\v{r}ej Du\\v{s}ek, Tom\\'a\\v{s} Nekvinda","submitted_at":"2020-08-03T10:43:30Z","abstract_excerpt":"We introduce an approach to multilingual speech synthesis which uses the meta-learning concept of contextual parameter generation and produces natural-sounding multilingual speech using more languages and less training data than previous approaches. Our model is based on Tacotron 2 with a fully convolutional input text encoder whose weights are predicted by a separate parameter generator network. To boost voice cloning, the model uses an adversarial speaker classifier with a gradient reversal layer that removes speaker-specific information from the encoder.\n  We arranged two experiments to com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.00768","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2008.00768/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":"2008.00768","created_at":"2026-07-05T01:24:10.893730+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.00768v1","created_at":"2026-07-05T01:24:10.893730+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.00768","created_at":"2026-07-05T01:24:10.893730+00:00"},{"alias_kind":"pith_short_12","alias_value":"672EZKVYEADT","created_at":"2026-07-05T01:24:10.893730+00:00"},{"alias_kind":"pith_short_16","alias_value":"672EZKVYEADTXCDJ","created_at":"2026-07-05T01:24:10.893730+00:00"},{"alias_kind":"pith_short_8","alias_value":"672EZKVY","created_at":"2026-07-05T01:24:10.893730+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02443","citing_title":"Breaking the Barriers of Text-Hungry and Audio-Deficient AI","ref_index":94,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX","json":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX.json","graph_json":"https://pith.science/api/pith-number/672EZKVYEADTXCDJECQ66ASFNX/graph.json","events_json":"https://pith.science/api/pith-number/672EZKVYEADTXCDJECQ66ASFNX/events.json","paper":"https://pith.science/paper/672EZKVY"},"agent_actions":{"view_html":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX","download_json":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX.json","view_paper":"https://pith.science/paper/672EZKVY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.00768&json=true","fetch_graph":"https://pith.science/api/pith-number/672EZKVYEADTXCDJECQ66ASFNX/graph.json","fetch_events":"https://pith.science/api/pith-number/672EZKVYEADTXCDJECQ66ASFNX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX/action/storage_attestation","attest_author":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX/action/author_attestation","sign_citation":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX/action/citation_signature","submit_replication":"https://pith.science/pith/672EZKVYEADTXCDJECQ66ASFNX/action/replication_record"}},"created_at":"2026-07-05T01:24:10.893730+00:00","updated_at":"2026-07-05T01:24:10.893730+00:00"}