Meta has introduced live voice conversations to Muse, its personal artificial intelligence agent, allowing users to interact with the assistant through speech rather than text.
The update also features an animated avatar that responds to conversations with facial expressions and gestures.
The feature is now available on the Muse app for mobile devices, the web and Mac computers, although access is currently limited to users in the United States. Users can initiate a voice conversation by tapping the phone icon in the top-right corner of the app.
During conversations, users can interrupt Muse at any time, with the AI responding immediately. The assistant can also continue working on tasks in the background while users engage in conversations, including when they are away from their devices.
The animated avatar complements the voice feature by displaying facial expressions and gestures that reflect the tone of the conversation. It also responds visually while users are speaking.
Users can customise how the avatar appears during conversations. They can opt for voice-only interactions, display the avatar alongside the chat or expand it to full screen. The feature also offers hundreds of voices with different personalities, which users can switch between at any time.
At its Connect conference, Meta also announced that users would be able to customise voices by describing their preferences, including changes to speaking speed and accent.
Meta said the voice feature could be used for activities such as brainstorming ideas, planning trips, preparing for job interviews and practising presentations.
The avatars are powered by Muse Realtime Avatar, an artificial intelligence model developed by Meta Superintelligence Labs, the company’s main AI unit. Meta first unveiled the technology at its Connect conference in September.
