Muse, Meta’s AI agent, can now hold voice conversations in real time. Its avatar animates as it speaks, adding a visible layer to an interaction that previously could happen through text.
The new features bring voice, avatar animation and voice customization together. Their rollout has begun in the US app, while Meta says support for other devices is coming.
Voice conversations with an animated face
During a voice chat, the avatar appears at the top of the conversation and moves in time with Muse’s words. The animation is meant to accompany the spoken response as the exchange unfolds.
Users can adjust the voice’s tone in two ways: ask Muse for a particular quality and let it choose, or browse options in the avatar settings. That gives people some control over how the agent sounds without requiring them to pick from settings alone.
The feature was tested in the iOS app. In brief use, Muse responded quickly over voice, but sometimes paused when it needed to look something up or check information. The same kind of delay can happen with a text query.
Personalization can make the experience feel unusual
The combination of a speaking voice and a moving avatar makes the AI feel more present in the conversation. It also means the agent’s appearance and animation can shape how the exchange comes across.
In the test described, the avatar was set to resemble Mark Zuckerberg as a reminder that the conversation was taking place with a Meta service. Watching that digital likeness speak felt strange, and the available voice could not be made to sound exactly like Zuckerberg.
That reaction points to a tradeoff built into expressive avatars. A face that moves with speech may make an agent easier to follow, while a lifelike or familiar appearance can also make the interaction feel uncanny. The source describes one person’s brief experience, rather than a broader user response.
Where Muse voice mode is headed
Meta says voice mode and expressive avatars are rolling out now in the US for the app. The company also says the features are coming “soon” to Meta’s AI glasses.
They are also planned for the forthcoming Muse Charm gadget, which is supposed to ship this year. This would extend Muse voice interactions beyond the app, although the source does not give a more specific release date for the glasses or gadget.
For now, the app rollout is the clearest way to try the new format. The announced device plans suggest Meta wants Muse to work across screens and wearables, but the voice and avatar experience described so far is the one available in the US app.
What changes in the conversation
Muse’s update is more than a change from typing to speaking: it pairs spoken answers with a visible, animated character and lets users influence the voice’s tone. That combination may make conversations feel more immediate, while pauses for information lookups remain part of the experience.
The feature’s next steps depend on its rollout to additional devices. Meta has named its AI glasses and Muse Charm as destinations, but the article gives no additional details about how voice mode will work on either one.