Google appears to be developing a more detailed voice customization system for Gemini. Instead of limiting users to a set of predefined voic...
Google appears to be developing a more detailed voice customization system for Gemini. Instead of limiting users to a set of predefined voices, the planned interface may let them adjust how Gemini sounds through four separate controls: Speed, Energy, Formality, and Warmth.
The feature was discovered by 9to5Google while examining code strings in the Google app 17.41.12 beta. Because the information comes from an APK teardown rather than an official product announcement, the controls should still be treated as an experimental feature that Google may change, delay, or decide not to release.
The beta code points to a new Customize section for Gemini voices. It includes four labeled characteristics:
Speed is expected to offer three levels: slow, normal, and fast. The other three controls—Energy, Formality, and Warmth—are expected to use low, medium, and high settings.
This would give users more control over Gemini’s delivery without requiring them to switch to a completely different preset voice. A user could keep the same general voice while changing its pace or conversational style to better suit the current task.
Gemini may gain more granular controls for how its voice sounds and feels.
Each setting is designed to affect a different part of the listening experience.
The Speed setting would control how quickly Gemini speaks. A slower pace may be easier to follow during explanations or language practice, while a faster pace could be more convenient for short updates and routine questions.
Energy would likely influence how lively or restrained Gemini sounds. A higher setting could make the assistant feel more animated, while a lower setting may produce a calmer delivery.
Formality could help users match Gemini’s voice to the situation. A less formal style may feel more natural for travel planning or casual brainstorming, while a more formal voice could be better suited to research, work, or professional discussions.
Warmth is expected to adjust how friendly and approachable the voice sounds. Increasing it could make everyday conversations feel more relaxed, while reducing it may result in a more neutral presentation.
These descriptions reflect the likely purpose of the controls based on their labels. Google has not yet published official documentation explaining exactly how each level will change the generated voice.
According to the strings examined by 9to5Google, a customized voice should be used in both Gemini Live and the regular Gemini chat experience when responses are read aloud.
That would be consistent with Gemini’s current voice setting. Google’s existing help documentation says that the selected Gemini voice applies across Gemini Apps, including Gemini Live and other features that provide spoken responses.
The important difference is that today’s official setting focuses on choosing a voice. The experimental interface would add another layer, allowing users to adjust the behavior and delivery of that voice.
The reported customization controls fit Google’s broader effort to make voice interaction feel more natural and locally relevant. Google has also discussed support for regional dialects, which could make Gemini’s spoken responses feel more familiar to users in different locations.
Voice localization is not only about changing an accent. Pacing, tone, warmth, and formality can all affect whether an assistant feels comfortable to use in a particular language, culture, or setting.
A voice suitable for a formal work discussion may feel unnecessarily stiff during a casual conversation. Similarly, a highly energetic delivery may work well for brainstorming but become tiring during a long research session. More granular controls could help address those differences.
Google is not the only company giving users more control over an AI assistant’s voice. Apple has announced that Siri AI in iOS 27 will allow users to customize pace and expressivity.
The two approaches are similar in direction, but Google’s experimental design appears to separate voice behavior into more categories. In addition to pace, Gemini may offer direct controls for energy, formality, and warmth.
This reflects a wider shift in voice assistants. The quality of a spoken AI experience is no longer determined only by whether the assistant understands a question correctly. The way it responds—how fast it speaks, how formal it sounds, and how emotionally appropriate the delivery feels—is becoming part of the product experience.
Describe your idea once, and We0 AI can generate a showcase site, pages, and CMS, then help you attract customers and traffic after launch.
One complete project generation for free registration
Best for trying one complete generation flow and seeing a first project draft quickly.
Voice interfaces are used in many different situations. Someone may speak with Gemini while driving, studying, planning a trip, reviewing a document, or working through a technical problem. A single fixed speaking style is unlikely to feel ideal in all of those contexts.
More detailed controls could provide several practical benefits:
The feature could also make voice assistants easier to use for longer conversations. Even a high-quality synthetic voice can become tiring when its pace or tone does not match the listener’s preference.
Google has not officially announced the four voice controls, and there is currently no confirmed release date. Their appearance in the Google app beta indicates active development, but APK teardowns cannot guarantee that a feature will reach the public version of the app.
Google may revise the labels, change the available levels, limit the feature to selected languages, or remove it before release. Availability could also vary by account, device, region, and Gemini language settings.
For now, users can continue changing Gemini’s preset voice through the Gemini mobile app:
The number of available voices varies by language, and Google notes that the voice-selection option is not available in every supported language.
The experimental interface includes Speed, Energy, Formality, and Warmth. Speed may use slow, normal, and fast levels, while the other controls may offer low, medium, and high settings.
Google has not officially launched them. They were found in the Google app 17.41.12 beta through an APK teardown, so the feature remains unconfirmed and may change before release.
The discovered code suggests that the customized voice will apply to Gemini Live and regular Gemini chats with spoken responses. This has not yet been confirmed in an official Google announcement.
Yes. Google currently allows users to select a preset Gemini voice in the mobile app, although the number of options varies by language and the setting is not available everywhere.
Google has not published an official definition for the experimental control. Based on the label, it is expected to influence how friendly, approachable, or emotionally warm Gemini sounds.
The planned Speed control is expected to include slow, normal, and fast options. Until the feature is released, users may not have a dedicated global speed slider in the standard Gemini voice settings.
More granular controls could help users adapt Gemini’s spoken responses to different tasks and personal preferences. They may also make long voice conversations feel more natural and less repetitive.
Apple has announced pace and expressivity controls for Siri AI in iOS 27. Gemini’s experimental system appears broader because it may also include separate settings for energy, formality, and warmth.
Google is testing a more flexible Gemini voice system built around four controls: Speed, Energy, Formality, and Warmth. The feature could let users adjust how Gemini sounds without replacing the selected voice entirely.
The controls were discovered in the Google app 17.41.12 beta and are expected to affect both Gemini Live and spoken replies in regular chats. However, Google has not officially announced the feature, so its final design and release timing remain uncertain.
If released, the update would move Gemini voice personalization beyond simple voice selection and toward task-specific, user-controlled speaking styles.
Start from one sentence and have a complete website in minutes.
