At the moment, we’re introducing Gemini 3.1 Flash TTS, the newest text-to-speech mannequin that delivers improved controllability, expressivity and high quality — empowering builders, enterprises and on a regular basis customers to construct the subsequent era of AI-speech functions.
Beginning at the moment, 3.1 Flash TTS is rolling out:
Improved speech high quality and controllability
We’ve improved the general speech high quality of Gemini 3.1 Flash TTS, making it our most pure and expressive mannequin up to now. On the Synthetic Evaluation TTS leaderboard, a benchmark that captures 1000’s of blind human preferences, 3.1 Flash TTS achieved a formidable Elo rating of 1,211.
