Google launched two new text-to-speech models: Gemini 3.8 Flash TTS and Flash-Lite TTS. They enable custom voice creation and line-by-line performance direction. Both models support over 100 languages and dialects. They include 2,000+ production-ready voices and voice replication from a 30-second sample. Built-in safety tools include SynthID watermarking and consent verification. The models are available in Google AI Studio, the Gemini API, and Gemini Notebook. Enterprise access is coming soon via API. They lead Hume AI’s Voice Design Benchmark and Overall Quality Index. Partners like Figma, HeyGen, and Agora are integrating them. Developers can test them in Google AI Studio’s audio playground.
What changed
Static preset voices with limited customization and control.
What it unlocks
Creating custom character voices and directing scene dialogue line by line with granular control.
- 71.4 Voice Design Benchmark score
- 60.8 accent modeling score
- 100+ languages supported
- 2,000+ production-ready voices
What you need to act on it
- access to Google AI Studio, Gemini API, or Gemini Notebook
- verbal consent recording for voice replication
Sources