nearly 400 packages in requirements.txt?

[GZY219]

The requirements.txt contains nearly 400 pacakges to install, only part of them are necessary if you just want to run the demo. I would like to provide a minimum environment setting.

system, cuda and python version

lsb_release -a
# Description:    Ubuntu 20.04.1 LTS
# Release:        20.04
nvcc --version
# Cuda compilation tools, release 12.1, V12.1.66
python --version
# Python 3.11.3

clone the repo, git lfs should be installed

https://github.com/baichuan-inc/Baichuan-Audio.git
cd Baichuan-Audio/web_demo
git lfs install
git clone https://huggingface.co/baichuan-inc/Baichuan-Audio-Base

install packages

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
pip install transformers==4.45.0
pip install diffusers==0.24.0
pip install huggingface-hub==0.25.1
pip install wheel
pip install packaging
pip install ninja==1.11.1.3
# download flash-attn whl from release, see https://github.com/Dao-AILab/flash-attention/releases
pip install flash_attn-2.7.4.post1+cu12torch2.5cxx11abiFALSE-cp311-cp311-linux_x86_64.whl
pip install omegaconf
pip install fire
pip install deepspeed
pip install decord
pip install opencv-python
pip install av
pip install cairosvg

run the demo

in web_demo/constants.py, MODEL_PATH is changed to MODEL_PATH = "Baichuan-Audio-Base". Then, just run the demo

python base_tts_demo.py

[Li-XYi]

Well done bro!