Skip to content

Apply diffusion models using the new Hugging Face diffusers package to synthesize music instead of images.

License

Notifications You must be signed in to change notification settings

timohromadka/audio-diffusion

 
 

Repository files navigation

Timo Hromadka MPhil Project.

This repository builds on the repository by Robert Dargawel Smith (https://github.com/teticio/audio-diffusion).

My additions to the repository include:

  • bug fixes
  • architecture modifications
  • conditional generation script, t5 model encoding
  • experimentation + hpc integration
  • result analysis
  • creative task experimentation (inpainting, outpainting, music style transfer) with novel models/datasets

Model Architecture

Model Architecture

About

Apply diffusion models using the new Hugging Face diffusers package to synthesize music instead of images.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 96.5%
  • Shell 3.0%
  • Python 0.5%