{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GVV5VQ4VXXFEJEDHUR644VNWMO","short_pith_number":"pith:GVV5VQ4V","schema_version":"1.0","canonical_sha256":"356bdac395bdca449067a47dce55b663a97f3ce11c83780ffb6cb4b9aa6461a9","source":{"kind":"arxiv","id":"2501.05787","version":1},"attestation_state":"computed","paper":{"title":"MARS6: A Small and Robust Hierarchical-Codec Text-to-Speech Model","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Akshat Prakash, Arnav Mehta, Elliott Dyson, Herman Kamper, Matthew Baas, Pieter Scholtz","submitted_at":"2025-01-10T08:41:42Z","abstract_excerpt":"Codec-based text-to-speech (TTS) models have shown impressive quality with zero-shot voice cloning abilities. However, they often struggle with more expressive references or complex text inputs. We present MARS6, a robust encoder-decoder transformer for rapid, expressive TTS. MARS6 is built on recent improvements in spoken language modelling. Utilizing a hierarchical setup for its decoder, new speech tokens are processed at a rate of only 12 Hz, enabling efficient modelling of long-form text while retaining reconstruction quality. We combine several recent training and inference techniques to "},"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":"2501.05787","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2025-01-10T08:41:42Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"e8914e5fa2e113da7725dd916d29d3ff1e7e84ba99d0c38966159d739f59d043","abstract_canon_sha256":"680df17ce50393bd8da84a85f5f0aae770c8d14a4bb5a226a0c2f75c7c8bc3bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:59:28.696891Z","signature_b64":"TRkyK6oHC1a+KBaV05/OyM9f434fHIyOKGYqqIQuCrVUucR9teEbA/sy+bDzP8krXBvLVGo12O25Xw6/cMAaDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"356bdac395bdca449067a47dce55b663a97f3ce11c83780ffb6cb4b9aa6461a9","last_reissued_at":"2026-07-05T09:59:28.696484Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:59:28.696484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MARS6: A Small and Robust Hierarchical-Codec Text-to-Speech Model","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"eess.AS","authors_text":"Akshat Prakash, Arnav Mehta, Elliott Dyson, Herman Kamper, Matthew Baas, Pieter Scholtz","submitted_at":"2025-01-10T08:41:42Z","abstract_excerpt":"Codec-based text-to-speech (TTS) models have shown impressive quality with zero-shot voice cloning abilities. However, they often struggle with more expressive references or complex text inputs. We present MARS6, a robust encoder-decoder transformer for rapid, expressive TTS. MARS6 is built on recent improvements in spoken language modelling. Utilizing a hierarchical setup for its decoder, new speech tokens are processed at a rate of only 12 Hz, enabling efficient modelling of long-form text while retaining reconstruction quality. We combine several recent training and inference techniques to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05787","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/2501.05787/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":"2501.05787","created_at":"2026-07-05T09:59:28.696544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.05787v1","created_at":"2026-07-05T09:59:28.696544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05787","created_at":"2026-07-05T09:59:28.696544+00:00"},{"alias_kind":"pith_short_12","alias_value":"GVV5VQ4VXXFE","created_at":"2026-07-05T09:59:28.696544+00:00"},{"alias_kind":"pith_short_16","alias_value":"GVV5VQ4VXXFEJEDH","created_at":"2026-07-05T09:59:28.696544+00:00"},{"alias_kind":"pith_short_8","alias_value":"GVV5VQ4V","created_at":"2026-07-05T09:59:28.696544+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.24496","citing_title":"Speech Token Prediction via Compressed-to-fine Language Modeling for Speech Generation","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO","json":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO.json","graph_json":"https://pith.science/api/pith-number/GVV5VQ4VXXFEJEDHUR644VNWMO/graph.json","events_json":"https://pith.science/api/pith-number/GVV5VQ4VXXFEJEDHUR644VNWMO/events.json","paper":"https://pith.science/paper/GVV5VQ4V"},"agent_actions":{"view_html":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO","download_json":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO.json","view_paper":"https://pith.science/paper/GVV5VQ4V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.05787&json=true","fetch_graph":"https://pith.science/api/pith-number/GVV5VQ4VXXFEJEDHUR644VNWMO/graph.json","fetch_events":"https://pith.science/api/pith-number/GVV5VQ4VXXFEJEDHUR644VNWMO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO/action/storage_attestation","attest_author":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO/action/author_attestation","sign_citation":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO/action/citation_signature","submit_replication":"https://pith.science/pith/GVV5VQ4VXXFEJEDHUR644VNWMO/action/replication_record"}},"created_at":"2026-07-05T09:59:28.696544+00:00","updated_at":"2026-07-05T09:59:28.696544+00:00"}