GoogleAudio model
Gemini 3.8 Flash TTS
Google's speech model that turns text into spoken audio, with 70 voices and a bill worked out from what the run used once the audio exists.
- Text to speech
From 5.4 credits per 1,000 characters, billed by use
Biggest gap
−23%
- XPlor3
- IDR 6,930per 1M tokens
- Official rate
- IDR 9,000per 1M tokens
Every other rate is in the table below.
Example output
No examples for this model yet, but you can try it in the studio right now.
Good for
- Narration for videos and ads in a single voice
- Two-speaker dialogue, for a podcast or a scene between characters
- Reference audio for the video models that take a voice
Specifications
- Reference material
- Text prompt only
- Tiers
- Flash, Flash_lite
- Prompt length
- up to 10,000 characters
Gemini 3.8 Flash TTS pricing
Each setting's rate beside the official price. Pick a credit package above the table to see roughly how much it buys.
Every rate is in rupiah per image, per video, per second, or per million tokens for speech. The XPlor3 column uses the base rate of Rp 99 a credit, so anyone on a bigger package pays less than what you see here. Official rates come from each provider's public price list, converted at Rp 18,000 to the dollar. The estimate column divides the package's credits by that row's rate and rounds down.
Run Gemini 3.8 Flash TTS now
The studio shows how many credits a render takes before you press Generate, so the bill never surprises you.