{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FQM4XTKBOK7262WZCVSLHRV6HE","short_pith_number":"pith:FQM4XTKB","schema_version":"1.0","canonical_sha256":"2c19cbcd4172bfaf6ad91564b3c6be392ff6e8655ab95892d413ed07fca70c56","source":{"kind":"arxiv","id":"2509.02020","version":2},"attestation_state":"computed","paper":{"title":"FireRedTTS-2: Towards Long Conversational Speech Generation for Podcast and Chatbot","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Feiyu Shen, Fenglong Xie, Junjie Li, Kun Xie, Xu Tang, Yao Hu","submitted_at":"2025-09-02T07:06:46Z","abstract_excerpt":"Current dialogue generation approaches typically require the complete dialogue text before synthesis and produce a single, inseparable speech containing all voices, making them unsuitable for interactive chat; moreover, they suffer from unstable synthesis, inaccurate speaker transitions, and incoherent prosody. In this work, we present FireRedTTS-2, a long-form streaming TTS system for multi-speaker dialogue generation, delivering stable, natural speech with reliable speaker switching and context-aware prosody. A new 12.5Hz streaming speech tokenizer accelerates training and inference, extends"},"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":"2509.02020","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-09-02T07:06:46Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"d4603213222801d02497bdb2f3ac23b7e3b3ca514272c6114df965fbb4b4ab29","abstract_canon_sha256":"0cd07dc72e5bfd507678c6690246cda449e601ad9923212051779f68c63806f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:40.890460Z","signature_b64":"Jn37bmsryYg7my1yJ9CGQJOrTNb512GlnIkUfh/gdYc2pyf2FlqNCg8IQgs6GLPhINhOEUGcpQYr4VPMCww9Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c19cbcd4172bfaf6ad91564b3c6be392ff6e8655ab95892d413ed07fca70c56","last_reissued_at":"2026-07-05T12:04:40.889976Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:40.889976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FireRedTTS-2: Towards Long Conversational Speech Generation for Podcast and Chatbot","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Feiyu Shen, Fenglong Xie, Junjie Li, Kun Xie, Xu Tang, Yao Hu","submitted_at":"2025-09-02T07:06:46Z","abstract_excerpt":"Current dialogue generation approaches typically require the complete dialogue text before synthesis and produce a single, inseparable speech containing all voices, making them unsuitable for interactive chat; moreover, they suffer from unstable synthesis, inaccurate speaker transitions, and incoherent prosody. In this work, we present FireRedTTS-2, a long-form streaming TTS system for multi-speaker dialogue generation, delivering stable, natural speech with reliable speaker switching and context-aware prosody. A new 12.5Hz streaming speech tokenizer accelerates training and inference, extends"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02020","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/2509.02020/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":"2509.02020","created_at":"2026-07-05T12:04:40.890026+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.02020v2","created_at":"2026-07-05T12:04:40.890026+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02020","created_at":"2026-07-05T12:04:40.890026+00:00"},{"alias_kind":"pith_short_12","alias_value":"FQM4XTKBOK72","created_at":"2026-07-05T12:04:40.890026+00:00"},{"alias_kind":"pith_short_16","alias_value":"FQM4XTKBOK7262WZ","created_at":"2026-07-05T12:04:40.890026+00:00"},{"alias_kind":"pith_short_8","alias_value":"FQM4XTKB","created_at":"2026-07-05T12:04:40.890026+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25369","citing_title":"Sarashina2.2-TTS: Tackling Kanji Polyphony in Japanese Speech Generation via Data Scaling and Targeted Data Synthesis","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09141","citing_title":"FlashTTS: Fast Streaming TTS with MTP Acceleration and X-pred Mean Flow Distillation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09019","citing_title":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20650","citing_title":"EmoInstruct-TTS: Dual-Path Instruction-Guided Emotional Speech Synthesis","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07080","citing_title":"dots.tts Technical Report","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06928","citing_title":"VoxCPM2 Technical Report","ref_index":35,"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":91,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30993","citing_title":"SwanVoice: Expressive Long-Form Zero-Shot Speech Synthesis for Both Monologue and Dialogue","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28249","citing_title":"HPRO: Hierarchical Progressive Reward Optimization via Preference Extraction for Emotional Text-to-Speech","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27258","citing_title":"PilotTTS: A Disciplined Modular Recipe for Competitive Speech Synthesis","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16026","citing_title":"From Flat Language Labels to Typological Priors: Structured Language Conditioning for Multilingual Speech-to-Speech Translation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16964","citing_title":"SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2601.15621","citing_title":"Qwen3-TTS Technical Report","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09386","citing_title":"Kinetic-Optimal Scheduling with Moment Correction for Metric-Induced Discrete Flow Matching in Zero-Shot Text-to-Speech","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22225","citing_title":"TTS-PRISM: A Perceptual Reasoning and Interpretable Speech Model for Fine-Grained Diagnosis","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11283","citing_title":"Multimodal Large Language Model-Enabled Video Translation: A Role-Oriented Survey","ref_index":88,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08363","citing_title":"CapTalk: Unified Voice Design for Single-Utterance and Dialogue Speech Generation","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE","json":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE.json","graph_json":"https://pith.science/api/pith-number/FQM4XTKBOK7262WZCVSLHRV6HE/graph.json","events_json":"https://pith.science/api/pith-number/FQM4XTKBOK7262WZCVSLHRV6HE/events.json","paper":"https://pith.science/paper/FQM4XTKB"},"agent_actions":{"view_html":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE","download_json":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE.json","view_paper":"https://pith.science/paper/FQM4XTKB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.02020&json=true","fetch_graph":"https://pith.science/api/pith-number/FQM4XTKBOK7262WZCVSLHRV6HE/graph.json","fetch_events":"https://pith.science/api/pith-number/FQM4XTKBOK7262WZCVSLHRV6HE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE/action/storage_attestation","attest_author":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE/action/author_attestation","sign_citation":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE/action/citation_signature","submit_replication":"https://pith.science/pith/FQM4XTKBOK7262WZCVSLHRV6HE/action/replication_record"}},"created_at":"2026-07-05T12:04:40.890026+00:00","updated_at":"2026-07-05T12:04:40.890026+00:00"}