Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yan, Canxiang, Jin, Chunxiang, Huang, Dawei, Yu, Haibing, Peng, Han, Zhan, Hui, Gao, Jie, Peng, Jing, Chen, Jingdong, Zhou, Jun, Ren, Kaimeng, Yang, Ming, Yang, Mingxue, Xu, Qiang, Zhao, Qin, Xiong, Ruijie, Lin, Shaoxiong, Wang, Xuezhi, Yuan, Yi, Wu, Yifei, Lyu, Yongjie, He, Zhengyu, Qiu, Zhihao, Fang, Zhiqiang, Huang, Ziyuan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.05516
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918191679143936
author Yan, Canxiang
Jin, Chunxiang
Huang, Dawei
Yu, Haibing
Peng, Han
Zhan, Hui
Gao, Jie
Peng, Jing
Chen, Jingdong
Zhou, Jun
Ren, Kaimeng
Yang, Ming
Yang, Mingxue
Xu, Qiang
Zhao, Qin
Xiong, Ruijie
Lin, Shaoxiong
Wang, Xuezhi
Yuan, Yi
Wu, Yifei
Lyu, Yongjie
He, Zhengyu
Qiu, Zhihao
Fang, Zhiqiang
Huang, Ziyuan
author_facet Yan, Canxiang
Jin, Chunxiang
Huang, Dawei
Yu, Haibing
Peng, Han
Zhan, Hui
Gao, Jie
Peng, Jing
Chen, Jingdong
Zhou, Jun
Ren, Kaimeng
Yang, Ming
Yang, Mingxue
Xu, Qiang
Zhao, Qin
Xiong, Ruijie
Lin, Shaoxiong
Wang, Xuezhi
Yuan, Yi
Wu, Yifei
Lyu, Yongjie
He, Zhengyu
Qiu, Zhihao
Fang, Zhiqiang
Huang, Ziyuan
contents Existing speech models suffer from competing requirements on token representations by understanding and generation tasks. This discrepancy in representation prevents speech language models from performing instruction-based free-form editing. To solve this challenge, we introduce a novel framework that unifies speech understanding, generation, and editing. The core of our unified model is a unified continuous speech tokenizer MingTok-Audio, the first continuous tokenizer to effectively integrate semantic and acoustic features, which makes it suitable for both understanding and generation tasks. Based on this unified continuous audio tokenizer, we developed the speech language model Ming-UniAudio, which achieved a balance between generation and understanding capabilities. Ming-UniAudio sets new state-of-the-art (SOTA) records on 8 out of 12 metrics on the ContextASR benchmark. Notably, for Chinese voice cloning, it achieves a highly competitive Seed-TTS-WER of 0.95. Leveraging this foundational model, we further trained a dedicated speech editing model Ming-UniAudio-Edit, the first speech language model that enables universal, free-form speech editing guided solely by natural language instructions, handling both semantic and acoustic modifications without timestamp condition. To rigorously assess the editing capability and establish a foundation for future research, we introduce Ming-Freeform-Audio-Edit, the first comprehensive benchmark tailored for instruction-based free-form speech editing, featuring diverse scenarios and evaluation dimensions spanning semantic correctness, acoustic quality, and instruction alignment. We open-sourced the continuous audio tokenizer, the unified foundational model, and the free-form instruction-based editing model to facilitate the development of unified audio understanding, generation, and manipulation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05516
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation
Yan, Canxiang
Jin, Chunxiang
Huang, Dawei
Yu, Haibing
Peng, Han
Zhan, Hui
Gao, Jie
Peng, Jing
Chen, Jingdong
Zhou, Jun
Ren, Kaimeng
Yang, Ming
Yang, Mingxue
Xu, Qiang
Zhao, Qin
Xiong, Ruijie
Lin, Shaoxiong
Wang, Xuezhi
Yuan, Yi
Wu, Yifei
Lyu, Yongjie
He, Zhengyu
Qiu, Zhihao
Fang, Zhiqiang
Huang, Ziyuan
Computation and Language
Artificial Intelligence
Sound
Audio and Speech Processing
Existing speech models suffer from competing requirements on token representations by understanding and generation tasks. This discrepancy in representation prevents speech language models from performing instruction-based free-form editing. To solve this challenge, we introduce a novel framework that unifies speech understanding, generation, and editing. The core of our unified model is a unified continuous speech tokenizer MingTok-Audio, the first continuous tokenizer to effectively integrate semantic and acoustic features, which makes it suitable for both understanding and generation tasks. Based on this unified continuous audio tokenizer, we developed the speech language model Ming-UniAudio, which achieved a balance between generation and understanding capabilities. Ming-UniAudio sets new state-of-the-art (SOTA) records on 8 out of 12 metrics on the ContextASR benchmark. Notably, for Chinese voice cloning, it achieves a highly competitive Seed-TTS-WER of 0.95. Leveraging this foundational model, we further trained a dedicated speech editing model Ming-UniAudio-Edit, the first speech language model that enables universal, free-form speech editing guided solely by natural language instructions, handling both semantic and acoustic modifications without timestamp condition. To rigorously assess the editing capability and establish a foundation for future research, we introduce Ming-Freeform-Audio-Edit, the first comprehensive benchmark tailored for instruction-based free-form speech editing, featuring diverse scenarios and evaluation dimensions spanning semantic correctness, acoustic quality, and instruction alignment. We open-sourced the continuous audio tokenizer, the unified foundational model, and the free-form instruction-based editing model to facilitate the development of unified audio understanding, generation, and manipulation.
title Ming-UniAudio: Speech LLM for Joint Understanding, Generation and Editing with Unified Representation
topic Computation and Language
Artificial Intelligence
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2511.05516