Google's MusicLM Can Generate Music from Anything You Type

We’ve known for quite a while that Google’s A.I. is one of the most advanced in the world. But creating music based on some random text? That’s some next-tier algorithm. And the music is not just some random, uninspiring rattle, either.  

“We introduce MusicLM, a model generating high-fidelity music from text descriptions such as ‘a calming violin melody backed by a distorted guitar riff,’” writes Google. 

“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.”

Pardon the French—Google’s AI engineers tend to speak like that. Essentially, just describe the music and Google MusicLM will generate the track accurately, based on your description. You can even paste descriptions of famous paintings and the AI will generate music with an eerily accurate mood and feel of that painting.

Take a look at the text descriptions below and listen to the music it generated through Google MusicLM. 

Google's new music model MusicLM is the breakthrough of the week.

Here it is in action.

Just describe the music and it'll generate the track: pic.twitter.com/xAhzHfGnMH

— Pete (@nonmayorpete) January 27, 2023

Google announces MusicLM: a model to generate music from text. Here are some crazy things it can do:

1. Given audio of a melody, it can generate new music inspired by that melody customized by prompts! Here's someone humming bella ciao turned into a cappella chorus, EDM, etc. pic.twitter.com/HKDnXI1C8U

— bleedingedge.ai (@bleedingedgeai) January 27, 2023

There's just one major catch, though: Google MusicLM is still in the experimentation labs and won't be released soon. However, if this were to fly, it would change the landscape of content creation and prevent copyright hits on millions of social media pages—all because of music.