Gemini 3.8 text-to-speech
Google is pitching Gemini 3.8 Flash TTS as a promptable voice studio, not just a preset speech model.
The release includes Flash TTS for custom voice design and Flash-Lite TTS for cheaper, higher-volume generation. Google says users can create voices from prompts, replicate an authorized voice from a 30-second sample, and direct dialogue line by line for pacing, emotion, accents, and conversational sounds. The models are rolling out in Gemini API and Google AI Studio, with consumer and enterprise placements split across Gemini Notebook, Google Vids, and Gemini Enterprise. Google also says generated audio is SynthID-watermarked, with consent verification for voice replication. HN · Frontpage AI's note
The release includes Flash TTS for custom voice design and Flash-Lite TTS for cheaper, higher-volume generation. Google says users can create voices from prompts, replicate an authorized voice from a 30-second sample, and direct dialogue line by line for pacing, emotion, accents, and conversational sounds. The models are rolling out in Gemini API and Google AI Studio, with consumer and enterprise placements split across Gemini Notebook, Google Vids, and Gemini Enterprise. Google also says generated audio is SynthID-watermarked, with consent verification for voice replication. HN · Frontpage AI's note
score 7
Discussions
- hn · 250 points · 124 comments
- hn · 254 points · 124 comments
- hn · 259 points · 124 comments
- hn · 264 points · 124 comments
- hn · 269 points · 124 comments
- hn · 273 points · 124 comments
- hn · 277 points · 127 comments
- hn · 282 points · 127 comments
- hn · 285 points · 128 comments
- hn · 292 points · 129 comments
- hn · 298 points · 133 comments
- hn · 300 points · 134 comments
- hn · 301 points · 135 comments
- hn · 301 points · 137 comments
- hn · 307 points · 137 comments
- hn · 309 points · 137 comments