requirements.txt

# --------- pytorch --------- #
torch>=2.3.0
torchvision>=0.15.0
lightning>=2.0.0
torchmetrics>=0.11.4

# --------- hydra --------- #
hydra-core==1.3.2
hydra-colorlog==1.2.0
hydra-optuna-sweeper==1.2.0

# --------- loggers --------- #
wandb

# --------- model --------- #
transformers[torch]>=4.34.0
peft

# --------- audiocraft --------- #
av
julius
einops
flashy
xformers
num2words
spacy
sentencepiece

# --------- evaluation --------- #
torchlibrosa
braceexpand
ftfy
progressbar
webdataset
wget
encodec
laion_clap --install-option="--no-deps"
frechet_audio_distance --install-option="--no-deps"  # transformers version conflict
git+https://github.com/haoheliu/audioldm_eval

# --------- data --------- #
resampy
mirdata
smart_open
git+https://github.com/moises-ai/moises-db.git

# --------- others --------- #
rootutils       # standardizing the project root setup
pre-commit      # hooks for applying linters on commit
rich            # beautiful text formatting in terminal
pytest          # tests