Voice conversations can be started by selecting the phone icon in the upper-right corner of the Muse app. Once connected, users can choose between an audio-only experience, a smaller animated character displayed alongside the conversation or a full-screen avatar.
The avatar is designed to make conversations feel more interactive by responding visually to both sides of an exchange. Its expressions and movements adjust to the conversation's tone, including while the user is speaking, rather than simply animating when Muse delivers a response.
Users can also personalize the assistant's appearance and voice. Meta offers hundreds of voice options with different characteristics and personalities, which can be changed during use. The company has also demonstrated a feature that lets users describe their preferred voice, including adjustments to speaking speed and accent.
The default avatar is a cartoon character called Jolly, although users can select other appearances, including people, animals and fictional characters. These customization options allow Muse to maintain a recognizable visual identity throughout a conversation.
The technology behind the animated characters is powered by Muse Realtime Avatar, a model developed by Meta Superintelligence Labs. The voice experience also incorporates Muse Realtime Voice, with both systems supporting the assistant's conversational capabilities.
Meta initially demonstrated the avatar technology at its Connect conference in September. In evaluations conducted by the company's own raters, Meta says its avatars were preferred over comparable offerings from Runway and HeyGen. However, the company has acknowledged that some demonstrations do not necessarily represent the exact avatar experience currently available in Muse.
The addition of voice and animated characters builds on Muse's initial launch in September. The assistant runs on Meta's Muse Spark model family, introduced in April, and was developed to handle tasks beyond conventional chatbot interactions.
Existing capabilities include purchasing products from retailers, composing and sending emails, and operating applications on Mac computers. The new voice interface allows users to communicate with Muse while those tasks continue, potentially making the assistant more convenient in situations where typing is impractical.
The update also introduces a more visible personality to Meta's AI agent. While competing assistants such as ChatGPT support spoken conversations, Meta is emphasizing the combination of responsive animation, customizable characters and voice interaction as a distinctive part of the Muse experience.
Meta chief AI officer Alexandr Wang highlighted the release with a demonstration on X featuring a personalized Muse character. In the example, the assistant appeared as a white dog and responded to Wang's comments about posting memes that could be mistaken for official Meta communications.
The broader release comes as Meta prepares to extend Muse beyond phones and computers. The company plans to bring voice conversations to its AI glasses and Muse Charm, a compact wearable device introduced at Connect.
Charm is designed to display the assistant's animated character on a small screen, allowing users to interact with Muse through spoken requests. Meta has indicated that the wearable is planned for release before the end of the year, although it has not provided a specific launch date. Timing for voice support on its AI glasses also remains unconfirmed.
The expansion is part of Meta's broader investment in artificial intelligence, which includes the development of its Hyperion data center in Louisiana and the work of Meta Superintelligence Labs.
For now, Muse's real-time voice and avatar capabilities are limited to the United States. Meta has not announced availability in additional markets or disclosed usage figures showing how widely its personal AI agent has been adopted.
The update gives Muse a more conversational interface while retaining its existing ability to carry out tasks. Its appeal will depend on whether users find the combination of voice interaction, animated characters and background task execution useful enough to incorporate into their everyday routines.
This analysis is based on reporting from BigGo Finance.
Image courtesy of Meta.
This article was generated with AI assistance and reviewed for accuracy and quality.
