Nvidia has created Fugatto, a new AI-powered music editor that can transform and combine sounds in ways not included in its training data. The system can merge animal sounds with musical instruments—like making trumpets bark or saxophones meow. Beyond these unusual combinations, Fugatto can modify human voices by changing accents and tonal qualities. The tool also offers music editing capabilities, letting users separate vocals from songs, add new instruments, or alter melodies. Nvidia hasn't announced when or if Fugatto will be available to the public.

Ad
Support our independent, free-access reporting. Any contribution helps and secures our future. Support now:
Bank transfer
Sources
Online journalist Matthias is the co-founder and publisher of THE DECODER. He believes that artificial intelligence will fundamentally change the relationship between humans and computers.
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.