# UniAudio 2.0: A Unified Audio Language Model with Text-Aligned Factorized Audio Tokenization

arXiv: https://arxiv.org/abs/2602.04683 · HTML: https://mlsanity.com/paper/2602.04683/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Feb 2026.

## Built on by (most cited first, 5 of 5)

- [Bagpiper: Solving Open-Ended Audio Tasks via Rich Captions](https://mlsanity.com/paper/2602.05220/) (2026)
- [FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation](https://mlsanity.com/paper/2608.24168/) (2026)
- [UAT: Unified Audio-Text Diffusion for Audio Generation, Editing, and Captioning](https://mlsanity.com/paper/2606.04939/) (2026)
- [Communicating Sound Through Natural Language](https://mlsanity.com/paper/2605.08750/) (2026)
- [TORUS: A Test of Rendering-Understanding Self-Coherence for Unified Audio Models](https://mlsanity.com/paper/2607.28896/) (2026)

## Abstract (arXiv)

We study two foundational problems in audio language models: (1) how to design an audio tokenizer that can serve as an intermediate representation for both understanding and generation; and (2) how to build an audio foundation model that generalizes in few-shot and zero-shot settings, analogous to large language models. To this end, we make the following two contributions. First, we propose ReasoningCodec, a discrete audio codec that factorizes audio into (i) reasoning tokens, which encode text-aligned, high-level analysis and planning representations for audio understanding and hierarchical generation, and (ii) reconstruction tokens, which encode semantic-rich acoustic cues for high-fidelity waveform reconstruction. This design achieves understanding performance comparable to strong continuous representations while improving generation quality and reconstruction fidelity over prior discrete tokenizers. Second, we introduce a unified autoregressive architecture for text and audio, together with multi-stage training and multi-task data construction. Using this framework, we train UniAudio 2.0 on 100B text tokens and 60B audio tokens. Across a wide range of speech, sound, and music tasks, UniAudio 2.0 performs competitively on in-domain evaluations and demonstrates strong few-shot and zero-shot generalization to unseen tasks. Demo, code, and checkpoints will be available at hrefhttps://dongchaoyang.top/UniAudio2Demo/https://dongchaoyang.top/UniAudio2Demo/.