Ingen beskrivning

Lengyue afc854ad85 Clean unused code & reorganize 2 år sedan
data_server 57755e1e74 Add rust data server 2 år sedan
fish_speech afc854ad85 Clean unused code & reorganize 2 år sedan
tools afc854ad85 Clean unused code & reorganize 2 år sedan
.dockerignore 57755e1e74 Add rust data server 2 år sedan
.gitignore 57755e1e74 Add rust data server 2 år sedan
.pre-commit-config.yaml f1ea577c4c [pre-commit.ci] pre-commit autoupdate 2 år sedan
LICENSE 62710e34a4 Add llama inference tool chain 2 år sedan
README.md 57755e1e74 Add rust data server 2 år sedan
dockerfile f8cfbb4ac0 Greatly improved data loading speed 2 år sedan
pyproject.toml 57755e1e74 Add rust data server 2 år sedan
pyrightconfig.json f7f2c03282 Support pytorch lightning 2 år sedan

README.md

Fish Speech

This repo is still under construction. Please check back later.

Setup

# Basic environment setup
conda create -n fish-speech python=3.10
conda activate fish-speech
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

# Install flash-attn (for linux)
pip3 install ninja && MAX_JOBS=4 pip3 install flash-attn --no-build-isolation

# Install fish-speech
pip3 install -e .

Rust Data Server

Since loading and shuffle the dataset is very slow and memory consuming, we use a rust server to load and shuffle the dataset. The server is based on GRPC and can be installed by

cd data_server
cargo build --release

Credits