Secondary Audio Stream over Screenshare (Second Mic/Application Share) #7
Description
Your use case
What would you like to do?
Add secondary audio source to pass unprocessed audio over "Screen Share"
Why would you like to do it?
I have talked to a lot of Discord users, and application audio share is highly sought after. I understand though that pulling system audio from a browser is hard if not impossible. A desktop client does fix this, however may be too far in the future to deal with right now.
By adding the ability to pass a secondary audio source through the screen share, without the noise cancelling applied, we could have an application audio sink to share the application audio provided one runs virtual audio cables. It's still more complicated than the desired way, however it would get around requiring system audio access as far as I know. It would just act as a secondary mic input.
How would you like to achieve it?
I am not a coder so I don't exactly know the limitations here. But I figure with the ability for Chrome browsers to share audio from tab over the Screen Share, this shouldn't be impossible to achieve.
Have you considered any alternatives?
Currently when I screen share over a voice call, I use Helvum to wire my Pipewire to the Mic input of Librewolf. I use Easyeffects to apply the filters over my mic for noise cancelling, gain control, gates, etc. I pipe the application post-effects to Librewolf, and apply the filtered one to the same inputs, so it goes all through my mic input.
While this works, for the average Discord user, this is needlessly complicated and pushes people away to Stoat or Fluxer.
Additional context
In the efforts of trying to move my communities to Matrix, application audio share has been a major thing brought up time and time again. It is a highly requested feature, and while Element Call doesn't seem to be jumping on this, now that SableCall is here, we could add this to improve the user experience for Discord migrants, especially now that we have some time to prepare before the dreaded ID push.
If there are ways to pull application audio directly, that would be preferred, but just having a dropdown to select a secondary mic source and push that over the screenshare stream that might be a good interim.
It has been suggested in Element Call's "Issues" to look at how Vencord/Vesktop does application audio capture and apply it. This might be a good way to implement this feature.