{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UC3YL543YVYZRSBWHLO6JPYOPE","short_pith_number":"pith:UC3YL543","schema_version":"1.0","canonical_sha256":"a0b785f79bc57198c8363adde4bf0e79163e03d9a7a0ca4146960df13a1d0502","source":{"kind":"arxiv","id":"2406.15752","version":1},"attestation_state":"computed","paper":{"title":"TacoLM: GaTed Attention Equipped Codec Language Model are Efficient Zero-Shot Text to Speech Synthesizers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"eess.AS","authors_text":"Guanrou Yang, Xiaofei Wang, Xie Chen, Yakun Song, Zhuo Chen, Ziyang Ma","submitted_at":"2024-06-22T06:39:52Z","abstract_excerpt":"Neural codec language model (LM) has demonstrated strong capability in zero-shot text-to-speech (TTS) synthesis. However, the codec LM often suffers from limitations in inference speed and stability, due to its auto-regressive nature and implicit alignment between text and audio. In this work, to handle these challenges, we introduce a new variant of neural codec LM, namely TacoLM. Specifically, TacoLM introduces a gated attention mechanism to improve the training and inference efficiency and reduce the model size. Meanwhile, an additional gated cross-attention layer is included for each decod"},"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":"2406.15752","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2024-06-22T06:39:52Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"607f218fc676441ae8bcc1e81715ccdba031649e303c92357bd749a507ca74d5","abstract_canon_sha256":"2334d4140db57e48de2f24e4404a65c8636c464a0eccad802d7a5ecd94a2d3d2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:35:27.042708Z","signature_b64":"QuEtqr3HBP0Nc4od9ZbDDFltXBVavQKdJX6OBJmRoTbzhk+Gml7/7rtMjhWY8KEbhqryx7ujrrO2qIp7o9rDCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0b785f79bc57198c8363adde4bf0e79163e03d9a7a0ca4146960df13a1d0502","last_reissued_at":"2026-07-05T08:35:27.042212Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:35:27.042212Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TacoLM: GaTed Attention Equipped Codec Language Model are Efficient Zero-Shot Text to Speech Synthesizers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"eess.AS","authors_text":"Guanrou Yang, Xiaofei Wang, Xie Chen, Yakun Song, Zhuo Chen, Ziyang Ma","submitted_at":"2024-06-22T06:39:52Z","abstract_excerpt":"Neural codec language model (LM) has demonstrated strong capability in zero-shot text-to-speech (TTS) synthesis. However, the codec LM often suffers from limitations in inference speed and stability, due to its auto-regressive nature and implicit alignment between text and audio. In this work, to handle these challenges, we introduce a new variant of neural codec LM, namely TacoLM. Specifically, TacoLM introduces a gated attention mechanism to improve the training and inference efficiency and reduce the model size. Meanwhile, an additional gated cross-attention layer is included for each decod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15752","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/2406.15752/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":"2406.15752","created_at":"2026-07-05T08:35:27.042276+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.15752v1","created_at":"2026-07-05T08:35:27.042276+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15752","created_at":"2026-07-05T08:35:27.042276+00:00"},{"alias_kind":"pith_short_12","alias_value":"UC3YL543YVYZ","created_at":"2026-07-05T08:35:27.042276+00:00"},{"alias_kind":"pith_short_16","alias_value":"UC3YL543YVYZRSBW","created_at":"2026-07-05T08:35:27.042276+00:00"},{"alias_kind":"pith_short_8","alias_value":"UC3YL543","created_at":"2026-07-05T08:35:27.042276+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.24794","citing_title":"V.O.I.C.E (Voice, Ownership, Identity, Control, Expression): Risk Taxonomy of Synthetic Voice Generation From Empirical Data","ref_index":78,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE","json":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE.json","graph_json":"https://pith.science/api/pith-number/UC3YL543YVYZRSBWHLO6JPYOPE/graph.json","events_json":"https://pith.science/api/pith-number/UC3YL543YVYZRSBWHLO6JPYOPE/events.json","paper":"https://pith.science/paper/UC3YL543"},"agent_actions":{"view_html":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE","download_json":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE.json","view_paper":"https://pith.science/paper/UC3YL543","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.15752&json=true","fetch_graph":"https://pith.science/api/pith-number/UC3YL543YVYZRSBWHLO6JPYOPE/graph.json","fetch_events":"https://pith.science/api/pith-number/UC3YL543YVYZRSBWHLO6JPYOPE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE/action/storage_attestation","attest_author":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE/action/author_attestation","sign_citation":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE/action/citation_signature","submit_replication":"https://pith.science/pith/UC3YL543YVYZRSBWHLO6JPYOPE/action/replication_record"}},"created_at":"2026-07-05T08:35:27.042276+00:00","updated_at":"2026-07-05T08:35:27.042276+00:00"}