Google recently announced that its researchers had created a “music-making” AI bot that generates original music from any text. Whether it’s a paragraph-long prompt or a one-word phrase, this new technology can transform it into melodic audio. The model system is called MusicLM.
So far, the function is not open to the public. The tech giant has released snippets of audio from songs created using short descriptions that can craft various genres, vibes, and even specific instruments.
MusicLM has been described as a “model generating high-fidelity music” from text descriptions like “a calming violin melody backed by a distorted guitar riff.” The model system can generate music at 24 kHz that remains consistent over several minutes,” according to a report by Cornell University.
The study also reveals that the model system can transform “whistled and hummed melodies” according to the specific style described in a text caption.
It’s unclear when the Google AI music maker will be released to the public, as it will likely stir conversations about intellectual property and copyright infringement. It will be interesting to see how the tech giant rolls out the platform in the near future.
Snippets of the AI-generated songs have been uploaded to Google’s Github account. You can listen here.