Google has introduced two new text-to-speech models, Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, designed to provide advanced creative control and scalable audio generation. The Flash TTS model focuses on deep character design and granular performance direction, allowing users to utilize natural language prompts to manipulate acting cues, pacing, and dialect shifts. It supports over 100 languages and provides access to a library of 2,000+ production-ready voices. For high-volume applications like dubbing and voice agents, the Flash-Lite TTS model offers cost-efficient, expressive scaling. Key technical capabilities include long-form generation with minimal speaker drift, native two-speaker scene staging, and the inclusion of non-verbal vocal bursts like laughs or sighs. The models also support voice replication from 30-second audio samples, protected by consent verification, C2CPA credentials, and SynthID watermarking. In performance benchmarks, Gemini 3.8 Flash TTS achieved the top spot on Hume AI’s Voice Design Benchmark with a score of 71.4 and led in accent modeling at 60.8. Furthermore, the models secured the first and second positions on Hume AI’s Overall Quality Index. These models are currently rolling out via the Gemini API and Google AI Studio for developers and enterprises.