Link: Nvidia claims a new AI audio generator can make sounds never heard before

Nvidia's new AI music editor, Fugatto, claims to produce unprecedented sounds, such as a meowing trumpet. This innovative tool synthesizes music and sounds from text and audio cues it hasn't previously encountered.

Fugatto can craft unique audio tracks based on imaginative prompts, showcasing its versatility in generating distinctive soundscapes. For example, it can merge a howling saxophone with electronic music and dog barks.

Among Fugatto's capabilities, it can also alter human voices, changing tones or accents to reflect different emotions like anger or calmness. Additionally, it can manipulate musical elements, isolating vocals or swapping instruments in a melody.

Although several AI audio tools exist from companies like Google and Adobe, Nvidia's Fugatto stands out by creating completely novel sounds. Yet, this space is not without controversy, as some AI-driven music tools have sparked copyright disputes.

Nvidia trained Fugatto on a vast array of sound data, including a significant collection from the BBC. Their researchers also enhanced the instructions to broaden the AI's capabilities without needing more data.

Nvidia has not confirmed a release date for Fugatto, leaving its availability uncertain. This raises intrigue and anticipation about when these groundbreaking soundscapes will become accessible to the public. #

--

Yoooo, this is a quick note on a link that made me go, WTF? Find all past links here.