Guides

Ready-to-go Dance Diffusion Music GPU Environment

The creators of this Dance Diffusion are Harmonai. We just provide a cloud environment that let's you run their project out of the box 🤙

This guide shows you how to spin up an Ubuntu Cloud GPU environment ready to generate music with Dance Diffusion.

  1. Sign up for an account
  2. It'll redirect you to an environment creation page pre-configured with the defaults you need. (We recommend sticking with the default GPU).
  3. At the bottom, add a payment method...we give you 30 minutes free but we just need to make sure people don't abuse our systems 🙂
  4. Hit create!

Open your new Dance Diffusion Brev environment:

brev open dance-diffusion --wait

Brev uses a CLI app to interface with and open your remote dev environments. If you don't have the Brev CLI install it here

Running the above command will open up a VSCode window SSHed into the remote machine. You'll first want to install the extensions you need to run the notebooks:

code --install-extension ms-toolsai.jupyter
code --install-extension ms-python.python

Generating music:

Now we're ready to go! There are two files in the repo to worry about: Dance_Diffusion.ipynb and Finetune_Dance_Diffusion.ipynb. Each notebook has instructions inside it on how to run it though each section should run out of the box. Jump into Dance_Diffusion.ipynb and start running!

Two things to consider:

  • The "Install dependencies and restart notebook" section will crash intentionally.
  • Make sure you run each Notebook with the Conda base environment kernel. This is where we install all Python dependencies at setup
  • If you regenerate your own sounds, the notebook will start a server running Gradio on port 7860. Because we are on a remote machine, we need to forward this port to our local machine: Next to the terminal icon in VSCode you should see a Ports section, forward port 7860 there and access it in your browser at localhost:7861

Message us on Discord if you have any issues!