A new open-source AI music generator drops
A new open-source AI music generator just landed on GitHub, turning simple text prompts and lyrics into complete vocal tracks. Early tests show clean stereo sound across styles like reggae and country, though pronunciation occasionally trips up. The model is now available to download and run locally.
This new open-source music generator turns simple text prompts and lyric sheets into full-length songs with vocals, all running right on your own machine.
What it can do
Feed it a style prompt—like a mid-tempo reggae groove with a melodic bassline or an acoustic country track with female vocals—paste your lyrics, and it spits out a finished track.
Early listens reveal remarkably clean audio output with solid stereo separation. That said, it isn't flawless: you will catch the occasional mispronunciation or slurred word. How it stacks up against current open-source benchmarks like Step 1.5 XL remains an open question until more stress tests roll in.
The project is live on GitHub right now, complete with model weights and local setup instructions.
Why it matters
If you want to sketch musical ideas without handing your data or monthly subscription fees to commercial platforms, another capable local model is a welcome arrival.
We think it is well worth grabbing for quick prototyping and demo tracks. The vocal quirks mean it is not ready to replace a final recording session, but having full creative control on your own hardware makes it worth tinkering with.
Sources (1)
- musicgen.com(opens in a new tab)Originalmusicgen.com