I bet most users just use the same device every time they want to load NoiseTorch (I do). The last update removed the nuisance of needing to type your password, but it can get even smoother. Maybe this could be a toggle switch under the Settings section.
Think of RTX voice, which uses the last device whenever it starts.

The repo is called polybar-pulseaudio-control but that's just because I use Polybar and I created the script for my own. But I'd like to rename it to just pulseaudio-control. The core of the script can be perfectly re-used for other status bars. I would introduce a --status-bar flag that sets the "compatibility mode" for that. If anyone wants to give it a try, leave a comment here!