Google researchers successfully created an AI-powered system that can turn any text into music. Dubbed as MusicLM, this sophisticated model can create high-quality music based on written descriptions like “a soothing acoustic guitar melody for relaxation”.

The Google research paper says MusicLM “casts the process of conditional music generation as a hierarchical sequence-to-sequence modeling task, and it generates music at 24 kHz that remains consistent over several minutes.”

Google is reportedly aware of the risks that could arise with the tool’s potential release and does not intend to do so as of the moment. In a paper, co-authors emphasized the “need for more future work in tackling these risks associated to music generation.”

It should be noted that Google isn’t the first to explore generative AI tools. Past attempts include OpenAI’s Jukebox, Riffusion, and Google’s own AudioML.

Google’s MusicLM offers features such as Audio Generation From Rich Captions, Story Mode, Long Generation, Text and Melody Conditioning and Painting Caption Conditioning. In addition, the AI can identify musicians’ levels of experience and expertise, places, accordion solos, epochs, generation diversity, and more.

Leave a comment

Your email address will not be published. Required fields are marked *