Google has created a new musical AI model ‘MusicLM’ that is
capable of turning text to music as well as whistled / hummed voices to
instrumental music.
The model generates music from text that describes a genre, vibe, or instrument. You can also check out a few samples that Google has shared of this incredible technology. In the audio compilation, you will hear music generated from the following descriptions: electronic song played in a videogame (0:00-0:15), meditation song played next to a river (0:15-0:30), fire (0:30-0:45), and fireworks (0:45-0:60).
Although the music doesn’t contain discernible lyrics, it
could still be enjoyed for its instrumental sounds. Above everything, it is
cool to be able to sort of create your own music simply by providing a computer
a prompt based on your mood or preference.