{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WIUPLPKWMHW4HEHUQWYLRF2NOU","short_pith_number":"pith:WIUPLPKW","schema_version":"1.0","canonical_sha256":"b228f5bd5661edc390f485b0b8974d752458c673ea8a96c0b6405508fea8d4ad","source":{"kind":"arxiv","id":"2402.08093","version":2},"attestation_state":"computed","paper":{"title":"BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.LG","authors_text":"Adam Michalski, Alexis Moinet, \\'Alvaro Mart\\'in-Cortinas, Ammar Abbas, Arent van Korlaar, Arnaud Joly, Bartosz Putrycz, Elena Sokolova, Ewa Muszy\\'nska, Fan Yang, Fatih Beyhan, Guillermo C\\'ambara, Haohan Guo, Kayeon Yoo, Mateusz {\\L}ajszczak, Soledad L\\'opez Gambino, Sri Karlapati, Thomas Drugman, Yang Li","submitted_at":"2024-02-12T22:21:30Z","abstract_excerpt":"We introduce a text-to-speech (TTS) model called BASE TTS, which stands for $\\textbf{B}$ig $\\textbf{A}$daptive $\\textbf{S}$treamable TTS with $\\textbf{E}$mergent abilities. BASE TTS is the largest TTS model to-date, trained on 100K hours of public domain speech data, achieving a new state-of-the-art in speech naturalness. It deploys a 1-billion-parameter autoregressive Transformer that converts raw texts into discrete codes (\"speechcodes\") followed by a convolution-based decoder which converts these speechcodes into waveforms in an incremental, streamable manner. Further, our speechcodes are b"},"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":"2402.08093","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T22:21:30Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"ce75d99e52fadac150f270602cbd853144941d4e9f76ee35ab2e7e659cd7be57","abstract_canon_sha256":"4af0ed086a670127dcb9977049b8767bbe7122eca488094c3537c55b5a705448"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:45:33.687913Z","signature_b64":"IKxQNnDZoJwVFtpG8WG8g/dWCiNMroubLdyUPDtTvbpMHT5C3+XAiTZELu7EqrcO/scYnOdbRnQvHp7DCdHxDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b228f5bd5661edc390f485b0b8974d752458c673ea8a96c0b6405508fea8d4ad","last_reissued_at":"2026-07-05T07:45:33.687050Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:45:33.687050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.LG","authors_text":"Adam Michalski, Alexis Moinet, \\'Alvaro Mart\\'in-Cortinas, Ammar Abbas, Arent van Korlaar, Arnaud Joly, Bartosz Putrycz, Elena Sokolova, Ewa Muszy\\'nska, Fan Yang, Fatih Beyhan, Guillermo C\\'ambara, Haohan Guo, Kayeon Yoo, Mateusz {\\L}ajszczak, Soledad L\\'opez Gambino, Sri Karlapati, Thomas Drugman, Yang Li","submitted_at":"2024-02-12T22:21:30Z","abstract_excerpt":"We introduce a text-to-speech (TTS) model called BASE TTS, which stands for $\\textbf{B}$ig $\\textbf{A}$daptive $\\textbf{S}$treamable TTS with $\\textbf{E}$mergent abilities. BASE TTS is the largest TTS model to-date, trained on 100K hours of public domain speech data, achieving a new state-of-the-art in speech naturalness. It deploys a 1-billion-parameter autoregressive Transformer that converts raw texts into discrete codes (\"speechcodes\") followed by a convolution-based decoder which converts these speechcodes into waveforms in an incremental, streamable manner. Further, our speechcodes are b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.08093","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/2402.08093/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":"2402.08093","created_at":"2026-07-05T07:45:33.687403+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.08093v2","created_at":"2026-07-05T07:45:33.687403+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.08093","created_at":"2026-07-05T07:45:33.687403+00:00"},{"alias_kind":"pith_short_12","alias_value":"WIUPLPKWMHW4","created_at":"2026-07-05T07:45:33.687403+00:00"},{"alias_kind":"pith_short_16","alias_value":"WIUPLPKWMHW4HEHU","created_at":"2026-07-05T07:45:33.687403+00:00"},{"alias_kind":"pith_short_8","alias_value":"WIUPLPKW","created_at":"2026-07-05T07:45:33.687403+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09295","citing_title":"N\\\"ushuVoice: Reviving the Voice of Endangered N\\\"ushu with Pitch-Aware Text-to-Speech","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03455","citing_title":"WavTTS: Towards High-Quality Zero-Shot TTS via Direct Raw Waveform Modeling","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31729","citing_title":"Is Natural Always Appropriate? Investigating Naturalness and Appropriateness Across Different Domains for TTS Evaluation","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31247","citing_title":"FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model","ref_index":112,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16964","citing_title":"SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2406.02430","citing_title":"Seed-TTS: A Family of High-Quality Versatile Speech Generation Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2412.10117","citing_title":"CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05611","citing_title":"X-Voice: Enabling Everyone to Speak 30 Languages via Zero-Shot Cross-Lingual Voice Cloning","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05611","citing_title":"X-Voice: Enabling Everyone to Speak 30 Languages via Zero-Shot Cross-Lingual Voice Cloning","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06327","citing_title":"A Novel Automatic Framework for Speaker Drift Detection in Synthesized Speech","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU","json":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU.json","graph_json":"https://pith.science/api/pith-number/WIUPLPKWMHW4HEHUQWYLRF2NOU/graph.json","events_json":"https://pith.science/api/pith-number/WIUPLPKWMHW4HEHUQWYLRF2NOU/events.json","paper":"https://pith.science/paper/WIUPLPKW"},"agent_actions":{"view_html":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU","download_json":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU.json","view_paper":"https://pith.science/paper/WIUPLPKW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.08093&json=true","fetch_graph":"https://pith.science/api/pith-number/WIUPLPKWMHW4HEHUQWYLRF2NOU/graph.json","fetch_events":"https://pith.science/api/pith-number/WIUPLPKWMHW4HEHUQWYLRF2NOU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU/action/storage_attestation","attest_author":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU/action/author_attestation","sign_citation":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU/action/citation_signature","submit_replication":"https://pith.science/pith/WIUPLPKWMHW4HEHUQWYLRF2NOU/action/replication_record"}},"created_at":"2026-07-05T07:45:33.687403+00:00","updated_at":"2026-07-05T07:45:33.687403+00:00"}