Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

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

Nov 25, 2024 - theverge.com
Nvidia has developed a new AI music editor named Fugatto, which can generate unique music, sounds, and speech using text and audio inputs it hasn't been trained on. The tool can create songs based on unusual prompts, produce sound effects based on descriptions, and alter the sound of a person's voice, including their accent and tone. It can also edit music by isolating vocals, adding instruments, and changing melodies. Fugatto was trained on a variety of datasets, including a library of sound effects from the BBC.

Despite the existence of other AI audio tools from companies like Stability AI, OpenAI, Google DeepMind, ElevenLabs, and Adobe, none claim to create entirely new and unheard-of sounds like Fugatto. However, Nvidia has not yet announced when or if the tool will be widely available. Some AI startups are facing copyright lawsuits over their music creation tools, and a recent report found that Nvidia and other companies trained AI models on subtitles from thousands of YouTube videos.

Key takeaways:

  • Nvidia has developed a new AI music editor, Fugatto, that can create unique sounds and music using text and audio inputs it hasn’t been trained on.
  • Fugatto can produce unique sound effects, transform the sound of someone’s voice, and edit music by isolating vocals, adding instruments, and changing melodies.
  • The AI tool was trained on a variety of datasets, including a library of sound effects from the BBC, and was built using a dataset with millions of audio samples.
  • While there are other AI audio tools available, Nvidia's Fugatto claims to create completely new and unheard-of sounds, but it's unclear when or if the tool will be widely available.
View Full Article

Comments (0)

Be the first to comment!