Google has introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, two text-to-speech models.

Flash TTS creates voices from text prompts and provides detailed control over how each line is delivered. Flash-Lite TTS is designed for high-volume, cost-efficient dubbing, audio content creation, and voice agents.

To replicate a voice, Google requires a 30-second recording and verifies the voice owner’s consent.

Google is starting to roll out both models in the Gemini API and Google AI Studio.

Claim check:

  • Google has introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS. (confirmed by the publication itself: evidence; «Today, we’re introducing two new text-to-speech models to the Gemini family»)
  • Gemini 3.8 Flash TTS creates voices from text prompts. (confirmed by the publication itself: evidence; «Create entirely new voices from scratch using natural language prompts»)
  • Gemini 3.8 Flash TTS provides detailed control over how each line is delivered. (confirmed by the publication itself: evidence; «Direct every performance line by line with granular control over acting cues, pacing, dialect shifts, and backchanneling.»)
  • Gemini 3.8 Flash-Lite TTS is optimized for high-volume, cost-efficient dubbing, audio content creation, and voice agents. (confirmed by the publication itself: evidence; «Built for high-volume, cost-efficient scale. Optimized for high-volume dubbing, audio content creation, and expressive voice agents»)
  • To replicate a voice, a 30-second recording and verification of the voice owner’s consent are required. (confirmed by the publication itself: evidence; «Recreate consistent vocal profiles from just a 30-second audio sample of your voice or a voice you have the rights to use, backed by built-in consent verification»)
  • Google is starting to roll out both models in the Gemini API and Google AI Studio. (confirmed by the publication itself: evidence; «Gemini 3.8 Flash TTS is rolling out starting today: For developers : In the Gemini API and Google AI Studio»)

Primary sources:

score 75.6 out of 100 · kind: announcement