Meta publishes Voicebox AI model: can generate audio reply messages for NPC conversations, etc.

CTOnews.com June 19 news, Meta has currently released the Voicebox AI model. Compared with the competitive model, which can only be replied with text or pictures, the advantage of the Voicebox AI model is mainly like its name, which can generate audio messages for reply.

▲ Voicebox AI model features, image source Meta it is reported that the Voicebox AI model only needs a 2-second audio sample to accurately identify audio details and timbre, and convert it into speech output based on text results, supporting English, French, German and Spanish. In addition, Voicebox also has the ability to "make up for the missing content based on the content before and after the voice clip".

The characteristic of ▲ Voicebox AI model, drawing source Meta

As a feature of the ▲ Voicebox AI model, the image source MetaMeta says that Voicebox can provide natural and real speech effects for AI-based virtual helpers or NPC in meta-universe. As for accessibility, Voicebox can also provide some assistance to people with damaged vocal cords.

After inquiry, CTOnews.com learned that the Voicebox AI model is still in the research and development stage. Meta said that it is aware of the potential harm of this artificial intelligence technology in terms of fake counterfeiting, so Meta is currently trying to find an effective way to distinguish between real voice and audio generated by Voicebox, which will not be made available to the public until a solution is found. You can now find more information about the Voicebox model here.