Google unveils Gemini 3.8 Flash TTS model, letting you design expressive audio experiences

Google has announced the release of Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, marking the latest expansion of its generative audio portfolio. The new models secured the first and second positions respectively on Hume AI's Overall Quality Index in blind human preference evaluations, outperforming competitors across six key global languages including Japanese, Brazilian Portuguese, Vietnamese, Arabic, Mexican Spanish, and Hindi. Supporting over 100 languages, these text-to-speech engines are now available through Google AI Studio, Gemini API, Gemini Enterprise, Gemini Notebook, and Google Vids, giving developers and enterprises immediate access to what Google describes as its “most expressive audio generation models yet”.

The flagship Gemini 3.8 Flash TTS targets deep creative applications, enabling users to generate entirely new character voices from natural language prompts for gaming, immersive audiobooks, podcasts, and interactive media. Its sibling model, Gemini 3.8 Flash-Lite TTS, is optimized for high-volume dubbing and voice agent deployments where cost-efficiency is paramount. Both models offer granular control over pacing, dialect shifts, and emotional nuance, along with native two-speaker scene staging that allows seamless multi-turn conversations from a single script. The platform includes an expansive library of 2,000+ production-ready voices and voice replication capabilities that recreate vocal profiles from just 30-second audio samples, backed by SynthID watermarking and C2PA credentials for consent verification and talent protection.