Gemini 3.8 Flash TTS and Flash-Lite TTS

Two new Google text-to-speech models that can design a custom voice from a written description or copy a real voice from a 30-second sample.

Flash TTS is the higher-quality model for creative direction and character voices; Flash-Lite TTS is the cheaper, faster one built for high volume like dubbing and voice agents
Describe a voice in plain language to create it, or pick from 2,000+ ready-made voices across 100+ languages and dialects
Voice cloning needs a 30-second sample plus a recorded spoken consent from the person whose voice it is, and isn't available in Illinois, Texas, the EEA, the UK, Switzerland, or India
All generated audio carries an inaudible watermark that marks it as AI-made
Available now in the Gemini API and Google AI Studio; Flash TTS also powers audio in Gemini Notebook and Flash-Lite TTS in Google Vids, with enterprise API access coming soon
Flash-Lite TTS replaces the earlier gemini-3.1-flash-tts-preview model

Producing voiceovers, audio versions of content, and voice agents without booking voice talent.

You need one consistent branded voice across many audio assets, or a lot of narration fast.

Content and video teams making podcasts, explainer videos, dubbed versions, or audio ads.

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.