---
license: mit
---

# Model Card for Make-An-Audio-2

<!-- Provide a quick summary of what the model is/does. -->

Make-An-Audio-2 is a text-conditional diffusion probabilistic model capable of generating high fidelity audio

## Model Details

Please refer our github repository for more details and how to use this model.

- **Repository:** https://github.com/bytedance/Make-An-Audio-2
- **Paper:** https://arxiv.org/abs/2305.18474