Overview
- Google began rolling out Gemini 3.8 Flash TTS and Gemini 3.8 Flash‑Lite TTS to developers through the Gemini API and Google AI Studio on September 23, 2026, with product appearances in Gemini Notebook and Google Vids and enterprise API access coming soon.
- Flash TTS is built for creative direction and long‑form performance, offering line‑by‑line acting cues, multi‑speaker scene staging, scripted nonverbal markers, and tools to keep voice timbre stable across hours of audio.
- Flash‑Lite TTS targets high‑volume use cases by trading some complexity for lower cost and faster production, making it suited for dubbing, automated agents, and large translation pipelines.
- The models support more than 100 languages and dialects and a library of over 2,000 production voices, and they can replicate a voice from roughly a 30‑second sample after explicit consent verification.
- Google published benchmark results placing Flash TTS at the top of vendor tests such as Hume AI and independent measures like the Artificial Analysis pronunciation benchmark, and partners including Figma, HeyGen, and others are beginning integrations while customers weigh cost, performance, and legal consent workflows.