MusicGen

Audio Generation

MusicGen is an advanced single language model (LM) for conditional music generation. The model, which was developed by Meta, works with compressed music tokens, eliminating the need to multiple models. It generates music based on text descriptions or melodies.