Google released two new Gemini 3.8 text-to-speech models (Flash and Flash-Lite) featuring over 2,000 voices and custom voice cloning from a 30-second sample. Developer Simon Willison built an interactive playground to test the API, highlighting its support for multi-speaker conversations with distinct voices and styles.
Background
Google has been expanding its Gemini family of AI models, including specialized text-to-speech capabilities. This release follows growing industry competition in neural voice synthesis, with companies like ElevenLabs and OpenAI also investing heavily in TTS technology.
- Source
- Simon Willison
- Published
- Sep 24, 2026 at 01:12 AM
- Score
- 7.0 / 10