Is there any way to change language processing model?

I’m really enjoying my Index so far, though I find it often misunderstands words I’ve said. I was wondering if there is a way (or if I can make a feature request) to change the language model it uses to interpret speech. I already have it set to Cloud with Local Fallback.

It would be nice to use a self hosted one or a paid model (e.g. Google just released Gemini 3.5 Transcribe) that can be hooked in for better speech recognition.

+1

In particular I’d like to be able to route to an Open WebUI instance. This makes it easy to configure models, tools, and any other of the many features included in Open WebUI with minimal configuration inside the Pebble app.