Nvidia on Monday showed off a new generative AI (genAI) model that can be used to create all kinds of sounds and music: Fugatto (which stands for Foundational Generative Audio Transformer Opus 1).
By entering a text prompt, a user can make Fugatto create basically any sound, such as a trumpet barking like a dog. The genAI model can also be used to change the dialect of a singer or turn a piano piece into a song, according to Reuters.
Fugatto has been trained on open source data, and there is currently no official release date. However, the idea is that the model will eventually be used in the production of music, films and games.