Step 1:

install jukeboxAI from official github

Official github

INSTALL COMMANDS

conda create --name jukebox python=3.7.5
conda activate jukebox
conda install mpi4py=3.0.3 # if this fails, try: pip install mpi4py==3.0.3
conda install pytorch=1.4 torchvision=0.5 cudatoolkit=10.0 -c pytorch
git clone https://github.com/openai/jukebox.git
cd jukebox
pip install -r requirements.txt
pip install -e .

conda install av=7.0.01 -c conda-forge 
pip install ./tensorboardX
pip install apex

Install trouble shoot

UnicodeEncodeError: 'ascii' codec can't encode character '\xe9' in position 13: ordinal not in range(128) PYTHONIOENCODING=UTF-8 pypa/pip#10219 SOLUTIONS:

• reduce sample_length_in_seconds
• reduce n_samples

torchvision=0.5.0 is not available through conda: Option 1: use pip Can not install torchvision 0.5.0 using conda Used pip install torchvision==0.5.0 instead of conda

torchvision=0.3 is available Option 2: use torchvision=0.3

ImportError: libffi.so.7: cannot open shared object file: No such file or directory sudo apt install libffi7 https://askubuntu.com/questions/1286772/libffi-so-7-cannot-open-shared-object-file-no-such-file-or-directory #libffi

RuntimeError: cublas runtime error : the GPU program failed to execute at /tmp/pip-req-build-pb3z3zl3/aten/src/THC/THCBlas.cu:315 Install newest versions

Step 2:

install local_jukeboxAI

Official github

git clone https://github.com/xiscoding/local_jukeboxAI.git

Step 3:

copy files from jukeboxAI into local_jukeboxAI

RUNNING THE CODE

runprompted.py

- Generates a continuation from sample audio file(s)
- uses a file selector gui 
- run and follow prompts