本地部署Fish-Speech语音识别工具指南

本地部署 fish-speech

1. 创建虚拟环境

conda create -n fish-speech python=3.10 -y
conda activate fish-speech

2. 安装依赖模块

pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --index-url https://download.pytorch.org/whl/cu121

3. 安装 fish-speech

cd /tmp
wget https://github.com/libsndfile/libsamplerate/releases/download/0.2.2/libsamplerate-0.2.2.tar.xz
tar -xvf libsamplerate-0.2.2.tar.xz
cd libsamplerate-0.2.2/
mkdir build
cd build
cmake ..
make

pip install -e .

4. 下载模型

python ./tools/download_models.py

5. 启动 fish-speech

# python fish_speech\webui\manage.py
python -m tools.webui     --llama-checkpoint-path "checkpoints/fish-speech-1.2"     --decoder-checkpoint-path "checkpoints/fish-speech-1.2/firefly-gan-vq-fsq-4x1024-42hz-generator.pth"     --decoder-config-name firefly_gan_vq

完结！

作者：engchina