Google unveils Flash TTS models for custom AI voices

google has introduced two new text-to-speech models, gemini 3.8 flash tts and flash-lite tts, which support over 100 languages. flash tts enables users to create new voices from text descriptions and includes a voice cloning feature that builds voice profiles from 30-second audio samples. the models allow for stage directions, two-voice dialogue, and nonverbal sounds. flash tts is aimed at creative projects like podcasts, audiobooks, and game characters, while flash-lite tts focuses on low-cost speech generation for dubbing, audio content, and voice agents. the models are rolling out through the gemini api and google ai studio, with gemini enterprise api access to follow.