Google isn’t a stranger to using AI to create music, having launched MusicLM in January to generate music from text. Google has now upped the ante, employing AI to read your brain and generate sound based on your brain activity.
Google’s AI Music
Google uses artificial intelligence (AI) to reconstruct music from brain activity captured by functional magnetic resonance imaging (fMRI) data.
Researchers examined fMRI data from five test subjects who listened to the same 15-second music clips from various genres such as blues, classical, country, disco, hip-hop, jazz, metal, pop, reggae, and rock.
The data was then used to train a deep neural network to learn about the relationship between brain activity patterns and musical elements such as rhythm and emotion.
Once trained, the model could use MusicLM to reconstruct music from an fMRI. Because MusicLM generates music from text, it was conditioned to produce music that was semantically similar to the original music stimuli.
When tested, the generated music matched the musical stimuli that the participant had initially listened to in terms of genre, instrumentation, mood, and other characteristics.
You can listen to several clips of the original music stimuli and compare them to the reconstructions generated by MusicLM on the research page’s website. The outcomes are astounding.
A 15-second clip of Britney Spears’ iconic “Oops!…I Did It Again” served as the stimulus for one clip. Like the original, the three reconstructions were poppy and upbeat.
The audio, of course, did not match the original because the study focuses on the different elements of music rather than the lyrical component.
Essentially, the model can read your mind (or, more precisely, your brain patterns) and generate music that is similar to what you were listening to.
To read our blog on “OpenAI CEO launches Worldcoin, the price spikes to 20%,” click here