Google launches Gemini 3.8 TTS models for custom, directed voices

Google introduced Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS on September 23, 2026, calling them its most expressive audio-generation models yet. [1] Both models are available across Google AI Studio, the Gemini API, Gemini Enterprise, Gemini Notebook, and Google Vids, where users can generate custom character voices and direct scene dialogue. [1] Google says the releases turn voice generation from static presets into a dynamic creative studio. [1]

Gemini 3.8 Flash TTS creates voices from natural-language prompts and supports line-by-line direction over acting cues, pacing, dialect shifts, and backchanneling. [1] Gemini 3.8 Flash-Lite TTS is optimized for high-volume dubbing, audio content creation, and expressive voice agents, with fine-grained control over tone, pacing, and expressive nuance. [1] Google says the models include safety features intended to support responsible use. [1]

The product split makes Google’s intended trade-off explicit: Flash emphasizes original voice and performance design, while Flash-Lite emphasizes volume and cost efficiency. Within the Gemini Audio family, the TTS pair sits alongside 3.5 Live Translate, 3.5 Transcribe, 3.8 Live, and 3.8 Live Extended Thinking, extending the portfolio into generated performances. Availability through several Google products also provides multiple distribution paths. What the sources don't address: how pricing, latency, quality benchmarks, consent controls, or voice-cloning safeguards compare across the two models.