Gemini 3.8 text-to-speech says hello (blog.google)

Google has unveiled two innovative text-to-speech models in the Gemini 3.8 release: Flash TTS and Flash-Lite TTS, which redefine voice generation by allowing creators and enterprises to craft dynamic and expressive audio experiences. The Flash TTS model emphasizes creative flexibility, enabling users to generate unique voices and control performances with meticulous detail, suitable for gaming, audiobooks, and more. In contrast, Flash-Lite TTS targets high-volume content creation, offering cost-effective solutions for dubbing and voice agents while maintaining fine control over tonal nuances. This advancement is significant for the AI/ML community as it enhances the capabilities of voice synthesis, providing support for over 100 languages and facilitating custom voice development with built-in safeguards for consent and transparency. Both models have demonstrated outstanding performance, ranking highly in voice design and quality benchmarks, which underscores their potential to improve user engagement in applications like podcasts and interactive media. With these tools, developers can easily create personalized audio experiences and are encouraged to experiment in the newly launched Google AI Studio.