MusiConGen

A Transformer-based text-to-music generation model

CommonProductMusicText-to-musicTransformer model
MusiConGen is a Transformer-based text-to-music generation model that enhances control over rhythm and chords through time conditioning. It is fine-tuned from the pre-trained MusicGen-melody framework. The model uses symbolic representations of chords and rhythm control, combined with five distinct styles of text descriptions to generate samples. The chords of the generated samples are estimated using the BTC chord recognition model, as detailed in the research paper.
Visit

MusiConGen Alternatives