Google introduces Gemini 3.8 Flash text-to-speech models
Google has introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, describing them as its most expressive audio-generation models so far. The models are designed for custom voice creation, scripted dialogue and multilingual audio experiences across Google AI Studio, the Gemini API, Gemini Notebook, Google Vids and, later, Gemini Enterprise.
Flash TTS focuses on designing original character voices and directing performances line by line, with controls for acting cues, pacing, dialect changes and conversational reactions. Flash-Lite TTS is optimized for high-volume, cost-efficient uses such as dubbing, audio production and voice agents. Google says Flash TTS supports voice design in more than 100 languages and dialects, while its library contains over 2,000 production-ready voices.
The models can stage two-speaker scenes, generate long-form audio and add cues such as laughter or sighs. Voice replication requires a 30-second sample and consent verification. Generated audio includes SynthID watermarking. Google says Flash TTS ranked first on Hume AI’s Voice Design Benchmark with a score of 71.4 and Flash-Lite ranked second on its Overall Quality Index. Both models are rolling out through the Gemini API and Google AI Studio; Flash TTS is also available in Gemini Notebook, while Flash-Lite is available in Google Vids.
Written by Sakhanda Wire AI from reporting by DeepMind. Facts, quotes and full context are in the original article below.