mywhisper-gtk: minimal Whisper voice input for Linux apps

[phplego]

Hi all,

I built a small Linux desktop app around Whisper for voice input into different applications, including terminal apps:
https://github.com/phplego/mywhisper

The workflow is simple:

• double Left Ctrl to start and stop recording
• Esc cancels the current recording
• the app sends audio for transcription and inserts the text into the active application

This is not an offline tool and it currently depends on an OpenAI API key set in the app settings.

I am sharing it here mostly because Ubuntu/X11 is the environment where I have actually used it.
The main thing I am trying to understand is how this kind of tool should work on Wayland.

On X11, global hotkeys and sending text to the active application are manageable.

On Wayland, the expected and acceptable way to do that is much less obvious.

If anyone here has experience shipping or using dictation tools on Ubuntu Wayland, I would be interested in pointers on:

• the right technical path for activation and text insertion (now it works via clipboard + send hardcoded [shift+Insert])
• whether an input-method-based approach is the right direction

I realize there are more mature dictation tools already, including local ones.

I am sharing this because the narrow use case may still be useful to someone, not because I think it replaces those projects.

Thank you in advance!