Google DeepMind has released its latest text-to-speech models, Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS. These models support over 100 languages, rank first in Hume AI's voice design benchmark, and hold the first and second positions, respectively, in the overall quality index. The new models allow users to generate custom character voices with precise control over accents and emotions on platforms such as Google AI Studio and the Gemini API. They also include built-in safety features, such as requiring verbal consent for voice replication and embedding SynthID watermarks in all AI-generated audio to prevent misinformation.