From the source
Google DeepMind released Voice remixing, which lets developers modify an existing prebuilt or designed voice with natural-language instructions to create a new, independent stored voice.
- The feature is reached through the POST /v1beta/voices endpoint with gemini-3.8-flash-tts, and it returns a voice_...
- ID that developers keep and reuse.
- Google AI Studio offers the same remixing interactively, so a voice can be shaped without writing the request by hand.
- Once a voice ID exists, speech can be synthesized with Gemini 3.8 Flash TTS or Gemini 3.8 Flash-Lite TTS.
- A Voice remixing guide is published for developers getting started.




