{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TX6SNWHMKGCDG5SLNVQEQ2MFO4","short_pith_number":"pith:TX6SNWHM","schema_version":"1.0","canonical_sha256":"9dfd26d8ec518433764b6d604869857703d4bff552720422ed15a99719750fec","source":{"kind":"arxiv","id":"2607.12496","version":1},"attestation_state":"computed","paper":{"title":"ZipL-Dialog: Memory-Efficient Long-Form Spoken Dialog Synthesis via Latent Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Jihwan Kim, Nam Soo Kim","submitted_at":"2026-07-14T08:28:48Z","abstract_excerpt":"Zero-shot dialog TTS benefits from flow-matching, but minute-scale generation on dense mel-spectrograms causes severe memory bottlenecks, often forcing unnatural chunked synthesis. We propose ZipL-Dialog, which shifts conditional flow-matching into a 4x time-compressed (25 Hz) latent space. To preserve acoustic fidelity under compression, we employ a deterministic mel autoencoder with auxiliary mel-domain supervision and optimize the ZipFormer's hierarchical downsampling schedule. Experiments show that ZipL-Dialog reduces maximum peak GPU memory by 11.22x and accelerates inference by 2.23x ove"},"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":"2607.12496","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-07-14T08:28:48Z","cross_cats_sorted":[],"title_canon_sha256":"15327de3d2cc09bb82e3f6e1202aaa2b7662717ed22db045bf1a11ac790cc86f","abstract_canon_sha256":"3d2f1cbc32fbb2bbd50caf9b5a1771674e979668174b528e6cb447f8fedee3ed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-15T01:21:38.095028Z","signature_b64":"dUT5nwUGxLPOyLbI+hdjQfqr9Erd/6uuQErx+wFt9Sd7P8Rgsq6CnqRhixrBbi+1P7mxDDwdd+z2A+ZHeyMOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dfd26d8ec518433764b6d604869857703d4bff552720422ed15a99719750fec","last_reissued_at":"2026-07-15T01:21:38.094087Z","signature_status":"signed_v1","first_computed_at":"2026-07-15T01:21:38.094087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ZipL-Dialog: Memory-Efficient Long-Form Spoken Dialog Synthesis via Latent Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"eess.AS","authors_text":"Jihwan Kim, Nam Soo Kim","submitted_at":"2026-07-14T08:28:48Z","abstract_excerpt":"Zero-shot dialog TTS benefits from flow-matching, but minute-scale generation on dense mel-spectrograms causes severe memory bottlenecks, often forcing unnatural chunked synthesis. We propose ZipL-Dialog, which shifts conditional flow-matching into a 4x time-compressed (25 Hz) latent space. To preserve acoustic fidelity under compression, we employ a deterministic mel autoencoder with auxiliary mel-domain supervision and optimize the ZipFormer's hierarchical downsampling schedule. Experiments show that ZipL-Dialog reduces maximum peak GPU memory by 11.22x and accelerates inference by 2.23x ove"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.12496","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/2607.12496/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":"2607.12496","created_at":"2026-07-15T01:21:38.094568+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.12496v1","created_at":"2026-07-15T01:21:38.094568+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.12496","created_at":"2026-07-15T01:21:38.094568+00:00"},{"alias_kind":"pith_short_12","alias_value":"TX6SNWHMKGCD","created_at":"2026-07-15T01:21:38.094568+00:00"},{"alias_kind":"pith_short_16","alias_value":"TX6SNWHMKGCDG5SL","created_at":"2026-07-15T01:21:38.094568+00:00"},{"alias_kind":"pith_short_8","alias_value":"TX6SNWHM","created_at":"2026-07-15T01:21:38.094568+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4","json":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4.json","graph_json":"https://pith.science/api/pith-number/TX6SNWHMKGCDG5SLNVQEQ2MFO4/graph.json","events_json":"https://pith.science/api/pith-number/TX6SNWHMKGCDG5SLNVQEQ2MFO4/events.json","paper":"https://pith.science/paper/TX6SNWHM"},"agent_actions":{"view_html":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4","download_json":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4.json","view_paper":"https://pith.science/paper/TX6SNWHM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.12496&json=true","fetch_graph":"https://pith.science/api/pith-number/TX6SNWHMKGCDG5SLNVQEQ2MFO4/graph.json","fetch_events":"https://pith.science/api/pith-number/TX6SNWHMKGCDG5SLNVQEQ2MFO4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4/action/storage_attestation","attest_author":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4/action/author_attestation","sign_citation":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4/action/citation_signature","submit_replication":"https://pith.science/pith/TX6SNWHMKGCDG5SLNVQEQ2MFO4/action/replication_record"}},"created_at":"2026-07-15T01:21:38.094568+00:00","updated_at":"2026-07-15T01:21:38.094568+00:00"}