Following its initial reveal at Connect 2026, Meta has officially begun rolling out real-time voice conversation features and expressive avatars within the "Muse" application. This step aims to transition the experience of dealing with artificial intelligence from mere written text to natural human interaction relying on voice and virtual body movements, and it is now available via smartphone applications, the web, and Mac devices.
Real-time Voice Interaction with Muse
The new feature allows users to think out loud with the smart assistant without the need to formulate precise text commands. Users can now tap the phone icon in the app to start an open dialogue, which facilitates brainstorming, trip planning, or even practicing for job interviews and presentations. The technical power here lies in the system's ability to process and understand speech as it flows, making responses feel immediate and interactive.
An Avatar That Expresses Emotions and Movements
Alongside the voice, Meta introduced Expressive Avatars technology that reacts visually to the tone of the conversation. The avatar uses precise facial expressions and gestures that match the context of the speech, giving the impression that the assistant is following the dialogue with interest. Meta also gave users complete flexibility in choosing how this avatar appears, whether through a voice-only interface, an avatar appearing alongside the text, or a full-screen mode for a face-to-face conversation experience.
Customizing the Voice Experience and Driving Technologies
This experience relies on "Muse Realtime Avatar" and "Muse Realtime Voice" technologies, which are designed to support smooth conversations that allow the user to interrupt the artificial intelligence in the middle of a sentence, just like in human dialogues. To add a personal touch, the company provided hundreds of diverse voice options, ranging from various local dialects to fictional characters, allowing the user to change the assistant's "persona" to suit their mood.
What Changes in the User Experience?
This update represents a shift from the traditional "question and answer" model to a "continuous collaboration" model. Instead of writing a specific request, the user can start with an incomplete idea and develop it through discussion with Muse while performing other tasks such as walking or cooking. This trend enhances the role of artificial intelligence as a digital companion capable of instantaneous response and understanding emotional context through the tone of voice and avatar expressions.
Availability and Future Plans
The roll-out of voice and avatar features has currently begun for Muse app users in the United States as a first step. Meta confirmed that these voice capabilities will soon arrive at AI Glasses and the "Muse Charm" device, indicating the company's desire to integrate its smart assistant into wearable devices to provide a usage experience that does not entirely rely on screens.