Audiocraft includes the training code and inference code for MusicGen, AudioGen, EnCodec, Multi Band Diffusion, MAGNeT, and AudioSeal. It also features a simple and controllable music generation LM with textual and melodic conditioning.
A state-of-the-art controllable text-to-music model.
A state-of-the-art text-to-sound model.
A state-of-the-art high fidelity neural audio codec.
An EnCodec compatible decoder using diffusion.
A state-of-the-art non-autoregressive model for text-to-music and text-to-sound.
Audio processing and generation with deep learning
Music generation with textual and melodic conditioning
Audio compression and decompression with EnCodec
Text-to-sound synthesis with AudioGen
Non-autoregressive text-to-music and text-to-sound synthesis with MAGNeT
Install Audiocraft using pip
Import Audiocraft in your Python script
Use the provided models and functions to process and generate audio
Train your own models using the provided training code
Experiment with different models and parameters to achieve the desired results