🛠️Gemini 3.8 Flash TTS: 2,000 Voices, 78s of Audio for 2.7¢
TL;DR
Simon Willison built a bring-your-own-key playground for Google's new Gemini 3.8 Flash TTS models and measured real costs. A 78-second two-speaker dialogue took about 20 seconds to generate and cost 2.74 cents.
Simon Willison built a bring-your-own-key playground for Google's new Gemini 3.8 Flash TTS models and measured real costs. A 78-second two-speaker dialogue took about 20 seconds to generate and cost 2.74 cents.
Key Points
Google shipped gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts on Sept 23
The library has 2,000+ voices, plus custom voice cloning from a 30-second sample you have rights to
The API defines multi-speaker conversations with per-line voice and delivery style instructions
Willison's playground runs in the browser thanks to the Gemini API's open CORS policy
Measured: 1m 18s of audio in ~20s for $0.0274 on the non-Lite model
Why It Matters
At under 3 cents a minute, synthetic multi-voice audio is cheap enough for podcast-style summaries of any document. Voice cloning from 30 seconds also raises the bar for consent checks.
Quick Facts
Comments
Be the first to comment
Enjoyed this article?
Get it daily. 7am. Free. Reads in 5 minutes.
Join 3,568 builders reading daily.