{"ID":2859284,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.05799","arxiv_id":"2510.05799","title":"Data-efficient Targeted Token-level Preference Optimization for LLM-based Text-to-Speech","abstract":"Aligning text-to-speech (TTS) system outputs with human feedback through preference optimization has been shown to effectively improve the robustness and naturalness of language model-based TTS models. Current approaches primarily require paired desirable and undesirable samples at the utterance level. However, such pairs are often limited in TTS output data, and utterance-level formulation prevents fine-grained token-level optimization needed for accurate pronunciation alignment. In this study, we propose TKTO that eliminates the need for paired data, enabling a more data-efficient training paradigm, and directly targets token-level units, automatically providing fine-grained alignment signals without token-level annotations. TKTO improves the challenging Japanese TTS accuracy by 39% and reduces CER by 54%, automatically assigning 12.8 times stronger reward to targeted tokens.","short_abstract":"Aligning text-to-speech (TTS) system outputs with human feedback through preference optimization has been shown to effectively improve the robustness and naturalness of language model-based TTS models. Current approaches primarily require paired desirable and undesirable samples at the utterance level. However, such pa...","url_abs":"https://arxiv.org/abs/2510.05799","url_pdf":"https://arxiv.org/pdf/2510.05799v2","authors":"[\"Rikuto Kotoge\",\"Yuichi Sasaki\"]","published":"2025-10-07T11:18:04Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\",\"cs.SD\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}