Gemini might finally allow you to decide on the tone of friendliness it conveys.

      Google has been working for several months to make Gemini sound more natural, expressive, and conversational. It now seems that the company is preparing to provide users with greater control over the AI's voice.

      Code discovered by Android Authority’s APK Insights in the latest beta version of the Google app indicates that Gemini may soon let users customize its voice through four distinct parameters: Energy, Formality, Warmth, and Speed. Rather than selecting from a predetermined list of personalities, users could modify these traits to create a voice that aligns more closely with their preferences.

      This feature hasn’t been officially announced yet, but its discovery suggests Google is moving towards making AI assistants more personalized.

      Gemini’s voice is becoming more adaptable

      The newly found controls were identified in the Google app version 17.41.12 beta, which includes references to a specific “Customize” section in Gemini’s voice settings.

      The code indicates that users will have the ability to adjust four traits:

      

      Energy: Low, Medium, High

      Formality: Low, Medium, High

      Warmth: Low, Medium, High

      Speed: Slow, Normal, Fast

      

      Rather than replacing Gemini’s current voices, these settings seem intended to enhance them. After customization, the personalized voice is expected to carry over to both Gemini Live and the regular chat experience, ensuring a consistent tone throughout the app.

      This represents a significant shift. Previously, choosing a Gemini voice mostly involved selecting from Google’s established personalities. The introduction of these new controls indicates that Google aims to empower users to shape the assistant’s tone rather than merely selecting one that feels most appropriate.

      Google recently updated Gemini's voices following I/O

      This discovery follows Google's update that revamped the voice selection experience after Google I/O.

      The redesigned picker has replaced the earlier carousel interface with a more streamlined list view and introduced two new voices: Flare and Glow, which take the place of the older Nova and Lyra options.

      The current voice lineup includes:

      Ursa

      Vega

      Pegasus

      Dipper

      Eclipse

      Capella

      Orbit

      Orion

      Flare

      Glow

      

      Notably, Google has also eliminated the descriptive labels that used to define each voice’s personality, such as “Calm” or “Bright.” Users now have to evaluate the voices by listening rather than relying on text-based descriptions.

      The update also brings a visual refresh to Gemini’s interface with sleeker, more modern icons for features like the microphone, camera, gallery, file uploads, video, screen sharing, and Gemini Live. These aesthetic changes are being deployed through a server-side update along with Gemini version 1.0.913571982.

      The timing is particularly interesting. At Google I/O, the company announced plans for regional dialects for Gemini. Voice customization aligns nicely with this larger initiative to make AI interactions feel less mechanical and more tailored to individual users.

      Google is not the only company moving in this direction. Apple’s iOS 27 introduces similar options for Siri, letting users adjust Pace and Expressivity, with those preferences reflecting across Siri-supported applications like Maps and Safari.

      Allowing users to control tone, warmth, and speaking style may seem like a minor modification, but it signifies a broader transformation in the AI landscape. Companies are no longer just competing on the capabilities of their assistants; they are increasingly competing on how those assistants sound while performing their tasks.