This is the server for the UltraEval-Audio project.

Setup

git clone https://github.com/UltraEval/Step-Audio.git
cd Step-Audio
conda create -n env python=3.10 -y
conda activate env
pip install -r requirments.txt

Run

python adv_api.py

Now, you can use --model step-voice in the UltraEval-Audio.