Small script to automate all (input+output i.e. speakers+microphone) audio recording.

[TheChymera]

So I've been banging my head against this for the longest time #131...
The wiki indeed explains it, but hard-coding the sink kept making my recordings miss either the input or the output as I over time got new speakers or new microphones.
Many important conversations lost to history and all that :D

So I put together a script which I hope people might find useful: https://github.com/TheChymera/quickscreenvid
If you want to include something similar into wf-recorder itself I think that would be a fabulous idea since “just record everything please” seems like a common and intuitive use case.

Some comments on the script:

1. The problem with just merging all sinks into one is it starts glitching. The linked script has a -b parameter to pass a regex for sink blacklisting, though that might be unintuitive for lots of people. I'm sure this could also be done “smarter” by checking which sinks have stuff playing on them, but that's beyond my current understanding of linux audio.
2. It would be really cool to actually record audio sinks separately (this would go beyond the scope of wf-recorder, or rather I used to have another script which just did that with other tools) and only merge them at the end. It's good to have them separately to fall back on if volumes don't work out in the combined in+out version. Currently they're as far as I can tell inseparable so volumes can't be adjusted later. On one hand, I foresee potential synchronization issues with that, though on the other hand I never encountered any.
3. Not sure how to make sure that the Combined sink is killed properly at the end without the additional script, especially not in a way that allows running this in the background (ideally you don't have a terminal window spewing stdout while recording your screen).