{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RBYBX64YB5GQP2I5THG63CF6LS","short_pith_number":"pith:RBYBX64Y","schema_version":"1.0","canonical_sha256":"88701bfb980f4d07e91d99cded88be5ca6b33e74844ac0a98cf09249d245a829","source":{"kind":"arxiv","id":"2404.03204","version":3},"attestation_state":"computed","paper":{"title":"RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Detai Xin, Dongchao Yang, Hiroshi Saruwatari, Jinyu Li, Kai Shen, Sheng Zhao, Shinnosuke Takamichi, Shujie Liu, Xu Tan, Yuancheng Wang, Zeqian Ju","submitted_at":"2024-04-04T05:15:07Z","abstract_excerpt":"We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis. While previous work based on large language models (LLMs) shows impressive performance on zero-shot TTS, such methods often suffer from poor robustness, such as unstable prosody (weird pitch and rhythm/duration) and a high word error rate (WER), due to the autoregressive prediction style of language models. The core idea behind RALL-E is chain-of-thought (CoT) prompting, which decomposes the task into simpler steps to enhance the robustness of LLM-based TTS. To accomplish this idea, RALL-E first predicts p"},"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":"2404.03204","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2024-04-04T05:15:07Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.SD"],"title_canon_sha256":"bd0d30525091e04b8b688aaf90223ce5424e2e5e4709787e15dee59006b4e96e","abstract_canon_sha256":"12a344504aa40ebf5ce834e03cf51653ba43912885b99f5054f70e00b767a2db"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:20:28.875440Z","signature_b64":"e5gEoEXX6D4Syj0ZeZnmALHWKNM0t7aa/kuahOhsmNd4j/h0SMzKlflxkM4Dc8xwUzvWqtF+txKEZ+IqQfe9Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88701bfb980f4d07e91d99cded88be5ca6b33e74844ac0a98cf09249d245a829","last_reissued_at":"2026-07-05T08:20:28.874910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:20:28.874910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RALL-E: Robust Codec Language Modeling with Chain-of-Thought Prompting for Text-to-Speech Synthesis","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Detai Xin, Dongchao Yang, Hiroshi Saruwatari, Jinyu Li, Kai Shen, Sheng Zhao, Shinnosuke Takamichi, Shujie Liu, Xu Tan, Yuancheng Wang, Zeqian Ju","submitted_at":"2024-04-04T05:15:07Z","abstract_excerpt":"We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis. While previous work based on large language models (LLMs) shows impressive performance on zero-shot TTS, such methods often suffer from poor robustness, such as unstable prosody (weird pitch and rhythm/duration) and a high word error rate (WER), due to the autoregressive prediction style of language models. The core idea behind RALL-E is chain-of-thought (CoT) prompting, which decomposes the task into simpler steps to enhance the robustness of LLM-based TTS. To accomplish this idea, RALL-E first predicts p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03204","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2404.03204/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":"2404.03204","created_at":"2026-07-05T08:20:28.874980+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.03204v3","created_at":"2026-07-05T08:20:28.874980+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03204","created_at":"2026-07-05T08:20:28.874980+00:00"},{"alias_kind":"pith_short_12","alias_value":"RBYBX64YB5GQ","created_at":"2026-07-05T08:20:28.874980+00:00"},{"alias_kind":"pith_short_16","alias_value":"RBYBX64YB5GQP2I5","created_at":"2026-07-05T08:20:28.874980+00:00"},{"alias_kind":"pith_short_8","alias_value":"RBYBX64Y","created_at":"2026-07-05T08:20:28.874980+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01677","citing_title":"UniVocal: Unified Speech-Singing Code-Switching Synthesis","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16964","citing_title":"SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2410.06885","citing_title":"F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching","ref_index":150,"is_internal_anchor":false},{"citing_arxiv_id":"2505.17589","citing_title":"CosyVoice 3: Towards In-the-wild Speech Generation via Scaling-up and Post-training","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2412.10117","citing_title":"CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS","json":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS.json","graph_json":"https://pith.science/api/pith-number/RBYBX64YB5GQP2I5THG63CF6LS/graph.json","events_json":"https://pith.science/api/pith-number/RBYBX64YB5GQP2I5THG63CF6LS/events.json","paper":"https://pith.science/paper/RBYBX64Y"},"agent_actions":{"view_html":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS","download_json":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS.json","view_paper":"https://pith.science/paper/RBYBX64Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.03204&json=true","fetch_graph":"https://pith.science/api/pith-number/RBYBX64YB5GQP2I5THG63CF6LS/graph.json","fetch_events":"https://pith.science/api/pith-number/RBYBX64YB5GQP2I5THG63CF6LS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS/action/storage_attestation","attest_author":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS/action/author_attestation","sign_citation":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS/action/citation_signature","submit_replication":"https://pith.science/pith/RBYBX64YB5GQP2I5THG63CF6LS/action/replication_record"}},"created_at":"2026-07-05T08:20:28.874980+00:00","updated_at":"2026-07-05T08:20:28.874980+00:00"}