Google has released a Brain2Music model that generates music based on human brain activity

CTOnews.com, July 26 (Xinhua) Google recently released a paper called Brain2Music, which uses artificial intelligence to read and analyze your functional magnetic resonance imaging (fMRI) data, and then automatically generate music based on your brain activity.

The researchers recruited five volunteers to listen to 15-second clips of blues, classical, country, disco, hip-hop, jazz, metal, pop, reggae and rock, and then recorded the volunteers' fMRI data.

Google uses the data to train a deep neural network to understand the relationship between brain activity patterns and different elements such as music rhythm and emotion.

Brain2Music is based on Google's previously launched MusicLM model, which generates music clips based on text, and generates 15-second music clips based on fMRI.

The results showed that one of the 15-second music clips generated by Brain2Music was basically the same as the original music heard by the test. I Did It Again.

CTOnews.com attaches the PDF download link to the paper here, which can be read in depth by interested users.