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

Meta unveils Audiobox, an AI that clones voices and generates ambient sounds

Dec 12, 2023 - venturebeat.com
Meta Platforms, the parent company of Facebook, Instagram, WhatsApp, and Oculus VR, has launched a free voice cloning program called Audiobox. The program, developed by researchers at the Facebook AI Research (FAIR) lab, can replicate a person's vocal stylings and generate voices and sound effects using voice inputs and natural language text prompts. Audiobox is built on a "family of models" for speech mimicry and ambient sound generation, all based on the self-supervised learning model Audiobox SSL.

However, Audiobox comes with restrictions. It cannot be used for commercial purposes and is not available to residents of Illinois or Texas due to state laws prohibiting the type of audio collection used in the demos. Unlike Meta's previous AI models, Audiobox is not open source. Despite these limitations, the rapid advancement of AI suggests that commercial versions of such technology may soon be available.

Key takeaways:

  • Meta Platforms, the parent company of Facebook, Instagram, WhatsApp, and Oculus VR, has released a free voice cloning program called Audiobox.
  • Audiobox can generate voices and sound effects using a combination of voice inputs and natural language text prompts. Users can type in a sentence for a cloned voice to say or describe a sound to generate.
  • Meta has created a "family of models" for Audiobox, one for speech mimicry and others for generating ambient sounds and sound effects. These models are built upon the self-supervised model Audiobox SSL.
  • Despite its capabilities, Audiobox cannot be used for commercial purposes and is restricted to those outside of the states of Illinois or Texas due to state laws prohibiting the type of audio collection Meta is doing for the demos.
View Full Article

Comments (0)

Be the first to comment!