←── back to feed
/topics/gemini-3-8-text-to-speech-models
Gemini 3.8 text-to-speech models
2 items●2 sources●updated 45d ago●trend 0
Google released Gemini 3.8 text-to-speech models capable of generating multi-voice conversations at low cost, with access to over 2,000 pre-built voices or custom voice cloning. The models enable applications like dialogue generation between distinct speakers.
- Over 2,000 pre-built voices available or users can clone custom voices
- Supports multi-voice conversation generation in single output
- Significantly reduced pricing compared to prior TTS offerings
- Released September 23, 2026
- Demonstrated with interactive playground UI and scripted dialogue examples