Large Language models (LLM) have demonstrated the capability to handle a variety of generative tasks. This paper presents the UniAudio system, which, unlike prior task-specific approaches, leverages LLM techniques to generate multiple types of audio (including speech, sounds, music, and singing) with given input conditions. UniAudio 1) first tokenizes all types of target audio along with other condition modalities, 2) concatenates source-target pair as a single sequence, and 3) performs next-token prediction using LLM. Also, a multi-scale Transformer model is proposed to handle the overly long sequences caused by the residual vector quantization based neural codec in tokenization. Training of UniAudio is scaled up to 165K hours of audio and 1B parameters, based on all generative tasks, aiming to obtain sufficient prior knowledge not only in the intrinsic properties of audio but also the inter-relationship between audio and other modalities. Therefore, the trained UniAudio model has the potential to become a foundation model for universal audio generation: it shows strong capability in all trained tasks and can seamlessly support new audio generation tasks after simple fine-tuning. Experiments demonstrate that UniAudio achieves state-of-the-art or at least competitive results on most of the 11 tasks. Demo and code are released at https://github.com/yangdongchao/UniAudio

翻译：大型语言模型（LLM）已展现出处理多种生成任务的能力。本文提出UniAudio系统，不同于以往针对特定任务的方法，该系统利用LLM技术，在给定输入条件下生成多种类型的音频（包括语音、声音、音乐和歌声）。UniAudio 1）首先将所有目标音频及其他条件模态进行分词化，2）将源-目标对拼接为单一序列，3）利用LLM执行下一词元预测。此外，针对分词过程中基于残差矢量量化的神经编解码器导致的超长序列问题，提出了一种多尺度Transformer模型。UniAudio的训练规模扩展至16.5万小时音频数据和10亿参数，基于所有生成任务，旨在获取充足的先验知识，不仅涵盖音频的内在属性，还包括音频与其他模态之间的相互关系。因此，训练后的UniAudio模型有潜力成为通用音频生成的基础模型：它在所有训练任务中展现出强大能力，并能在简单微调后无缝支持新的音频生成任务。实验表明，UniAudio在11项任务中的大多数上取得了最先进或至少具有竞争力的结果。演示和代码已发布在https://github.com/yangdongchao/UniAudio。