teticio/audio-diffusion
Audio generation model using diffusion architecture to synthesize music from noise.
Not currently ranked — collecting fresh signals.
star history
This project adapts image-based diffusion models to generate audio waveforms instead. It leverages the Hugging Face diffusers library to train and run diffusion models that produce music. The repository provides a Gradio web interface for audio generation and includes pretrained latent audio diffusion models. Users can generate audio samples by sampling from the diffusion process, with support for conditional generation based on text or audio encodings.
Frequently asked
- What is teticio/audio-diffusion?
- Audio generation model using diffusion architecture to synthesize music from noise.
- Is audio-diffusion open source?
- Yes — teticio/audio-diffusion is open source, released under the GPL-3.0 license.
- What language is audio-diffusion written in?
- teticio/audio-diffusion is primarily written in Jupyter Notebook.
- How popular is audio-diffusion?
- teticio/audio-diffusion has 792 stars on GitHub.
- Where can I find audio-diffusion?
- teticio/audio-diffusion is on GitHub at https://github.com/teticio/audio-diffusion.