Sam Witteveen · 2026-09-24 · notable
Sam Witteveen — 'Gemini 3.8 Flash TTS with Voice Cloning'
Sam Witteveen's 24 September 2026 video covers Gemini 3.8 Flash TTS, Google's speech model released a day earlier, and its voice cloning from a 30-second sample with the speaker's recorded consent.
Sam Witteveen looks at the voice-cloning side of Google's new Gemini 3.8 Flash TTS model.
What is it?
'Gemini 3.8 Flash TTS with Voice Cloning' went up on Sam Witteveen's channel on 24 September 2026, one day after Google released the model. Gemini 3.8 Flash TTS is Google's new text-to-speech model with over 100 languages and more than 2,000 ready-made voices.
How does it work?
Voice replication in Google's model needs a 30-second audio sample plus a verbal consent recording from the voice owner that matches the reference speaker. Designed and replicated voices can be saved and reused, so a persona sounds the same across calls. Replication is not offered in Illinois, Texas, the EEA, the UK, Switzerland or India.
Why does it matter?
Voice cloning now sits inside the same Gemini API developers use for text, instead of a separate vendor. A walkthrough from a hands-on creator is a quick way to judge output quality and the consent step before building on it.
Who is it for?
developers building voice apps