I’ve been using Dragon natural speech for at least a decade and it finally died on me last night. Some sort of file corruption, I spent hours trying to fix it and trying to enable Windows native speech text. But it didn’t work. I edit I edited so many registries.

[(https://github.com/Melvynx/Parler?tab=readme-ov-file)]

It’s called Handy STT It’s very minimal and clean and it works on Linux And I can actually say Linux and have it transcribe it correctly. Completely local, has an ungodly amount of language models to choose from, works in multiple languages, The dock isn’t massive and disappears when you hit a hotkey, Compact neatly into the system tray, It does voice recording if you’re into that and it’s open source. I’ve actually put Claude to work forking a release version. there’s only like two or three things I would change about this. So I can’t say enough good things about it. It saved me somewhere between 150 to 180 bucks for a replacement copy of Dragon.

Also, most importantly, it doesn’t spaz out and add twelve of the same character or make you use a box that you have to copy from in order to transfer the text.

As for the changes that I mentioned, and I’m completely nitpicking here, but since I’ve been using Dragon for ten years, I gotten used to it kind of typing your text as you speak. Even though obviously I can’t process it that fast. It just feels weird to me. The other thing would be Text to Speech integration period And maybe being able to move the little dock around. But the more I use it, the more I think that’s a good spot for it.