Gemini 3.1 Flash TTS – with directed prompts • Hacker News Board

Gemini 3.1 Flash TTS – with directed prompts

#1

Post by aanet » Wed, Apr 15, 2026, 5:47 PM UTC

Gemini 3.1 Flash TTS – with directed prompts
simonwillison.net

Re: Gemini 3.1 Flash TTS – with directed prompts

#2

Post by aanet » Wed, Apr 15, 2026, 5:47 PM UTC

The 3 examples, with three distinct styles, are fascinating.

I'd like to see one with cockney accent, just for lulz

Re: Gemini 3.1 Flash TTS – with directed prompts

#3

Post by roscas » Wed, Apr 15, 2026, 6:53 PM UTC

Hope they release an offline model for Ollama, a small one easy to work with for TTS in other languages.

Re: Gemini 3.1 Flash TTS – with directed prompts

#4

Post by xmichael909 » Wed, Apr 15, 2026, 9:45 PM UTC

roscas wrote: post #3

Hope they release an offline model for Ollama, a small one easy to work with for TTS in other languages.

So a free model? Tons of other people doing similar with amazing results https://huggingface.co/spaces/Inferless/Open-Source-TTS-Gall...

Re: Gemini 3.1 Flash TTS – with directed prompts

#5

Post by Insensitivity » Wed, Apr 15, 2026, 10:56 PM UTC

No matter what I wrote in the audio profile, AI Studio never followed it, regardless of scene or context.

For example, I tried to get a male voice and kept getting female ones. Not sure if it's an AI Studio bug or I was doing something wrong.

Re: Gemini 3.1 Flash TTS – with directed prompts

#6

Post by voxic11 » Wed, Apr 15, 2026, 11:25 PM UTC

Insensitivity wrote: post #5

No matter what I wrote in the audio profile, AI Studio never followed it, regardless of scene or context. For example, I tried to get a male voice and kept getting female ones. Not sure if it's an AI Studio bug or I was doing something wrong.

voice is determined by the voice parameter, you can't control it via the prompt, the prompt only directs how the chosen voices delivers the lines.