{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T15:20:56Z","timestamp":1775229656666,"version":"3.50.1"},"reference-count":54,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10890510","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T17:15:02Z","timestamp":1741799702000},"page":"1-5","source":"Crossref","is-referenced-by-count":3,"title":["Preference Alignment Improves Language Model-Based TTS"],"prefix":"10.1109","author":[{"given":"Jinchuan","family":"Tian","sequence":"first","affiliation":[{"name":"Carnegie Mellon University,Language Technologies Institute"}]},{"given":"Chunlei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tencent AI LAB"}]},{"given":"Jiatong","family":"Shi","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University,Language Technologies Institute"}]},{"given":"Hao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tencent AI LAB"}]},{"given":"Jianwei","family":"Yu","sequence":"additional","affiliation":[{"name":"Tencent AI LAB"}]},{"given":"Shinji","family":"Watanabe","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University,Language Technologies Institute"}]},{"given":"Dong","family":"Yu","sequence":"additional","affiliation":[{"name":"Tencent AI LAB"}]}],"member":"263","reference":[{"key":"ref1","article-title":"A survey on neural speech synthesis","author":"Tan","year":"2021"},{"key":"ref2","article-title":"Fastspeech: Fast, robust and controllable text to speech","volume-title":"NeurPIS","volume":"32","author":"Ren"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1452"},{"key":"ref4","first-page":"5530","article-title":"Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech","volume-title":"ICML","author":"Kim"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/taslp.2021.3129994"},{"key":"ref6","article-title":"High fidelity neural audio compression","author":"D\u00e9fossez","year":"2023","journal-title":"Transactions on Machine Learning Research"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447523"},{"key":"ref8","article-title":"Speechtokenizer: Unified speech tokenizer for speech language models","volume-title":"ICLR","author":"Zhang"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/SLT61566.2024.10832289"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1878"},{"key":"ref11","article-title":"Neural codec language models are zero-shot text to speech synthesizers","author":"Wang","year":"2023"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00618"},{"key":"ref13","article-title":"Uniaudio: Towards universal audio generation with large language models","volume-title":"ICML","author":"Yang"},{"key":"ref14","article-title":"Seed-tts: A family of high-quality versatile speech generation models","author":"Anastassiou","year":"2024"},{"key":"ref15","article-title":"Speak foreign languages with your own voice: Cross-lingual neural codec language modeling","author":"Zhang","year":"2023"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3288409"},{"key":"ref17","article-title":"Audiopalm: A large language model that can speak and listen","author":"Rubenstein","year":"2023"},{"key":"ref18","article-title":"Soundstorm: Efficient parallel audio generation","author":"Borsos","year":"2023"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447112"},{"key":"ref20","article-title":"Towards audio language modeling-an overview","author":"Wu","year":"2024"},{"key":"ref21","article-title":"A survey of reinforcement learning from human feedback","author":"Kaufmann","year":"2023"},{"key":"ref22","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017"},{"key":"ref23","first-page":"27730","article-title":"Training language models to follow instructions with human feedback","volume-title":"NeurIPS","volume":"35","author":"Ouyang"},{"key":"ref24","article-title":"Direct preference optimization: Your language model is secretly a reward model","author":"Rafailov","year":"2023","journal-title":"NeurPIS"},{"key":"ref25","article-title":"Model alignment as prospect theoretic optimization","volume-title":"ICML","author":"Ethayarajh"},{"key":"ref26","article-title":"SimPO: Simple preference optimization with a reference-free reward","author":"Meng","year":"2024"},{"key":"ref27","first-page":"4447","article-title":"A general theoretical paradigm to understand learning from human preferences","volume-title":"AISTATS","author":"Azar"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.626"},{"key":"ref29","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023"},{"key":"ref30","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref31","article-title":"Speechalign: Aligning speech generation to human preferences","author":"Zhang","year":"2024"},{"key":"ref32","article-title":"Enhancing zero-shot text-to-speech synthesis with human feedback","author":"Chen","year":"2024"},{"key":"ref33","article-title":"Robust zero-shot text-to-speech synthesis with reverse inference optimization","author":"Hu","year":"2024"},{"key":"ref34","article-title":"Tango 2: Aligning diffusion-based text-to-audio generative models through direct preference optimization","author":"Majumder","year":"2024","journal-title":"ACM Multimedia"},{"key":"ref35","article-title":"Simple and controllable music generation","volume-title":"NeurIPS","volume":"36","author":"Copet"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.405"},{"key":"ref37","article-title":"Fine-tuning language models with advantage-induced policy alignment","author":"Zhu","year":"2023"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.2307\/2334029"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1965"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2826"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447759"},{"key":"ref43","article-title":"Decoupled weight decay regularization","volume-title":"ICLR","author":"Loshchilov"},{"key":"ref44","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","volume-title":"ICML","author":"Radford"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1982"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1345"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-439"},{"key":"ref48","author":"Yamagishi","year":"2019","journal-title":"CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit"},{"key":"ref49","article-title":"Rall-e: Robust codec language modeling with chain-of-thought prompting for text-to-speech synthesis","author":"Xin","year":"2024"},{"key":"ref50","first-page":"2709","article-title":"Yourtts: Towards zero-shot multi-speaker tts and zero-shot voice conversion for everyone","volume-title":"ICML","author":"Casanova"},{"key":"ref51","article-title":"Some things are more cringe than others: Preference optimization with the pairwise cringe loss","author":"Xu","year":"2023"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1938"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2650"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/ojsp.2024.3378602"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Hyderabad, India","start":{"date-parts":[[2025,4,6]]},"end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10890510.pdf?arnumber=10890510","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T05:22:32Z","timestamp":1774416152000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10890510\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":54,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10890510","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}