GaronG

I'm having the same issue, I submitted mine on Oct 4th and still nothing.

I wonder if dynamically changing the temp based on the requested tone + a small "how to write skill" could produce more human varied writing, instead of using the same temp for the full request.

I've used a few and got so invested that I built my own. If I had to suggest some, they are also all Open Source:

Handy - It's great, supports Mac, Win and Linux. It's very focused on dictation, but it doesn't go too far outside of that. It's also completely free https://handy.computer/

VoiceInk - Mac only, and pretty good if you're on a Mac. It starts at $25 for one device. https://tryvoiceink.com/

Glimpse - the one I built. Mac and Windows, free local dictation. Some other features are paid, w/ a 2 week free trial. https://tryglimpse.cc

How's code signing and notarization on macOS with the embedded Python? That's been the painful part every time I've shipped something with a bundled runtime.

Also, can you mix Rust commands and Python in the same app?

At this point Google is just AI search, it's fast but even now I still have issues with it's AI overview making up small details. I'd rather wait a bit longer for a better LLM to synthesize results than have to fact check Google.