GoogleapiVoice Remixing for Gemini 3.8 TTS
For developers on the Gemini API and in Google AI Studio: adjust an existing AI voice with plain-language instructions and save the result as a new voice.
Details
- You describe the change in words, such as age, timbre, pitch, pacing, vocal energy, or accent, and the model applies it to the base voice
- Each remix returns a new voice ID and an audio preview, and that voice can be remixed again to keep refining it
- Base voices can be any of the 30 prebuilt voices, a voice from the extended library, or a stored voice your project already owns
- A remix is a standalone voice, so deleting the original does not affect it
- Remixed voices work with Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS
- Requires a paid Gemini API project, allows up to 200 stored voices per project, and keeps a voice for one year from its last use
Best for
Getting a consistent brand voice for narration or voice agents without designing one from scratch.
When to use it
Start here if a preset voice is close to what you need but the pace, pitch, or accent is off.
Who benefits
Developers and audio or video teams producing voiceover at volume through the API.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →