Google launched Gemini 3.8 Flash TTS and Flash-Lite TTS with generative voice design from natural language prompts across 100+ languages, 2,000+ preset voices, and voice replication from 30-second samples with spoken consent verification.
All generated audio includes SynthID watermarking; replicated voices also receive C2PA credentials. Google claims #1 ranking on Hume AI’s Voice Design Benchmark. Models available via Gemini API and Google AI Studio.
Voice replication is restricted in Illinois, Texas, EEA, UK, Switzerland, and India. Voice remixing feature coming soon.