{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VQ342OQISRN7JVARDPOGQTYLHS","short_pith_number":"pith:VQ342OQI","schema_version":"1.0","canonical_sha256":"ac37cd3a08945bf4d4111bdc684f0b3c972741d76a6b039b45f2308f6f3655cc","source":{"kind":"arxiv","id":"2506.13053","version":3},"attestation_state":"computed","paper":{"title":"ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Daniel Povey, Fangjun Kuang, Han Zhu, Liyong Guo, Long Lin, Weiji Zhuang, Wei Kang, Zengwei Yao, Zhaoqing Li","submitted_at":"2025-06-16T02:48:17Z","abstract_excerpt":"Existing large-scale zero-shot text-to-speech (TTS) models deliver high speech quality but suffer from slow inference speeds due to massive parameters. To address this issue, this paper introduces ZipVoice, a high-quality flow-matching-based zero-shot TTS model with a compact model size and fast inference speed. Key designs include: 1) a Zipformer-based vector field estimator to maintain adequate modeling capabilities under constrained size; 2) Average upsampling-based initial speech-text alignment and Zipformer-based text encoder to improve speech intelligibility; 3) A flow distillation metho"},"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":"2506.13053","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2025-06-16T02:48:17Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"2e4e4aff975b9b72b7295de67ac608f71119800c114342f20d7743c8b80f26da","abstract_canon_sha256":"247caa43930de93a6b3b294d2361fd93c20c11b9842e997b52b1a8e43b40f460"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:55.661263Z","signature_b64":"R1CVvVZmWqwoDxi2FGbUJUkF9/sRXBLiiXBFbi3QaU5T/PI+ragC+0koOTjvSpwaQDwXSVe0ZK5FZf6iCVshAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac37cd3a08945bf4d4111bdc684f0b3c972741d76a6b039b45f2308f6f3655cc","last_reissued_at":"2026-07-05T11:49:55.660776Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:55.660776Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Daniel Povey, Fangjun Kuang, Han Zhu, Liyong Guo, Long Lin, Weiji Zhuang, Wei Kang, Zengwei Yao, Zhaoqing Li","submitted_at":"2025-06-16T02:48:17Z","abstract_excerpt":"Existing large-scale zero-shot text-to-speech (TTS) models deliver high speech quality but suffer from slow inference speeds due to massive parameters. To address this issue, this paper introduces ZipVoice, a high-quality flow-matching-based zero-shot TTS model with a compact model size and fast inference speed. Key designs include: 1) a Zipformer-based vector field estimator to maintain adequate modeling capabilities under constrained size; 2) Average upsampling-based initial speech-text alignment and Zipformer-based text encoder to improve speech intelligibility; 3) A flow distillation metho"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13053","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/2506.13053/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":"2506.13053","created_at":"2026-07-05T11:49:55.660837+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.13053v3","created_at":"2026-07-05T11:49:55.660837+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13053","created_at":"2026-07-05T11:49:55.660837+00:00"},{"alias_kind":"pith_short_12","alias_value":"VQ342OQISRN7","created_at":"2026-07-05T11:49:55.660837+00:00"},{"alias_kind":"pith_short_16","alias_value":"VQ342OQISRN7JVAR","created_at":"2026-07-05T11:49:55.660837+00:00"},{"alias_kind":"pith_short_8","alias_value":"VQ342OQI","created_at":"2026-07-05T11:49:55.660837+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09234","citing_title":"End-to-End Training for Discrete Token LLM based TTS System","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07015","citing_title":"Towards Unified Song Generation and Singing Voice Conversion with Accompaniment Co-Generation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06928","citing_title":"VoxCPM2 Technical Report","ref_index":41,"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":108,"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":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08729","citing_title":"Unison: Harmonizing Motion, Speech, and Sound for Human-Centric Audio-Video Generation","ref_index":50,"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":43,"is_internal_anchor":false},{"citing_arxiv_id":"2507.09318","citing_title":"ZipVoice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00688","citing_title":"OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08729","citing_title":"Unison: Harmonizing Motion, Speech, and Sound for Human-Centric Audio-Video Generation","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19679","citing_title":"MMControl: Unified Multi-Modal Control for Joint Audio-Video Generation","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS","json":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS.json","graph_json":"https://pith.science/api/pith-number/VQ342OQISRN7JVARDPOGQTYLHS/graph.json","events_json":"https://pith.science/api/pith-number/VQ342OQISRN7JVARDPOGQTYLHS/events.json","paper":"https://pith.science/paper/VQ342OQI"},"agent_actions":{"view_html":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS","download_json":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS.json","view_paper":"https://pith.science/paper/VQ342OQI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.13053&json=true","fetch_graph":"https://pith.science/api/pith-number/VQ342OQISRN7JVARDPOGQTYLHS/graph.json","fetch_events":"https://pith.science/api/pith-number/VQ342OQISRN7JVARDPOGQTYLHS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS/action/storage_attestation","attest_author":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS/action/author_attestation","sign_citation":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS/action/citation_signature","submit_replication":"https://pith.science/pith/VQ342OQISRN7JVARDPOGQTYLHS/action/replication_record"}},"created_at":"2026-07-05T11:49:55.660837+00:00","updated_at":"2026-07-05T11:49:55.660837+00:00"}