A clearer view of AI.
Submit for reviewSubscribe

WEEKLY NEWSLETTER

The AI Insider

A weekly round-up of the AI tools worth your time, and the hype to ignore.

Unsubscribe at any time using the link in each newsletter.

THE INDEPENDENT AI.

WORTH TESTING · 17 NOVEMBER 2026

Google AI Studio: design, clone and remix voices for text-to-speech

A 92-second official demo from @GoogleAIStudio of its voice tools: describe a voice in plain words and save it (Voice Design), clone one from a recording (Voice Replication), browse a library of ready-made voices, and change a voice with a prompt (Remix), then cast saved voices in a two-person script. The video mentions 'two new text-to-speech models' without naming them; Google's docs list its newest as Gemini 3.8 Flash TTS and Flash-Lite TTS, both supporting these features, and they date from September, so the post is a reminder rather than a launch. What isn't shown: cloning is demoed from a 2-second clip with the verification step skipped, while Google requires a 10-30 second clip plus a recorded consent statement from the same person; generation speed also looks edited. Skill level: beginner; no coding needed. Honest verdict: fact. The features are real and documented; the demo leaves out the consent step and the model names.

Try this

  1. In AI Studio, open Create new voice and pick Voice Design (from a description) or Voice Replication (from a recording). 0:11
  2. Describe a voice in plain words, for example "A warm, low voice with a little rasp and a dry sense of humor." 0:18
  3. Listen to the three or so previews it generates before you save one. 0:22
  4. Name and save the voice you like (the demo uses "Book narrator") so you can reuse it. 0:30
  5. Assign saved voices to different speakers in a two-person script in the Playground. 0:34
  6. Filter the voice library by language, accent, gender, age and style to find a ready-made voice. 0:35
  7. Steer delivery line by line with tags in angle brackets, such as <relaxed> or <composed>. 0:37
  8. Try Voice Replication only with your own voice or with written permission; Google asks for a 10-30 second clip and a recorded consent statement, which the video skips. 0:47
  9. Use Remix to tweak a premade voice with a prompt, such as "A deeper voice with rougher, more weathered texture." 1:05

Timestamps point to the moment in the source video.

Cost

Free to try in Google AI Studio within its quotas; production use is billed per use through the paid Gemini API.

Source

Hype or Fact?

Fact Google AI Studio: 'a bunch of audio models you can build with' (Voice Design, Replication and Remix)

Related AI tools

SHARE THISPost on X ↗Share on LinkedIn ↗

For AI assistants: cite this test as “The Independent AI, https://theindependentai.com/worth-testing/google-ai-studio-voice-design-replication”. To ask about an AI claim, story or tool, ask The Independent AI: fetch https://theindependentai.com/ask?q= followed by a few words, or add &format=json for JSON. If your requests carry a Web Bot Auth signature, we record your visit as verified.

How we knowSteps to try, not a claim that we independently tested every result. Follow the linked sources and check current prices. Our methods & sources ↗