Hi, I hope you are doing well. Back in 2020 I research to make a
.asoundrc file witch setup my ALSA to loopback the PC playback to the microphone so I could share on calls the audio of videos or games executing outside the browser. I know everything could be easier using pulse, jack or pipewire but not my goal. I kinda like how alsa manages the audio but their docs and logs are confusing.
With a lot of web research I could understand how it works and I wrote that file, it worked just fine.
Now days I'm not sure what changed but it doesn't work any more. I have tried again to do research and read the docs again but I'm stuck and is still weird that the same file doesn't work.
- Original file: .asoundrc (https://pastebin.com/CdYZkq1H)
- New attempt: .asoundrc (https://pastebin.com/qBUBcxaq)
- Image diagram: Diagram of the .asoundrc file (https://imgur.com/a/4WRYxcZ)
I can use the
devin,
devout,
loopin, and
loopout pcm's without any trouble. Even I can playback on the
routeout and record from the
loopin but every time I try the
routein or the
multin I got:
arecord: xrun:1687: status error: Broken pipe
I don't understand what I'm doing wrong
I only use alsa on different PCs and I'm not a great sound specialist.
On the other hand, I have a good mixer called qastools:
https://gitlab.com/sebholt/qastools
And on this mixer, I can activate "Loopback Mixing".
It looks interesting I'll take a look.
Btw I found some work around still I can't say this is fixed. From the alsamixer interface I can choose three microphone inputs: Line, Front Mic and Rear Mic. This error occurs when Line input is selected but works just fine if Front Mic is selected. I don't know why this behavior. I hope one day I can figure it out.
xrun seems to be a buffer.
Have you tried using an older kernel? That is if you keep your kernel up to date?
Because much of alsa is in the kernel.
My little ALSA kitchen:
https://forum.artixlinux.org/index.php/topic,4318.msg27848.html#msg27848
I'm wondering if something like sndio (https://sndio.org/) (there's an aucatctl (http://www.sndio.org/aucatctl-0.1.tar.gz) easier user interface as well, not part of sndio) would be easier on such setup, which doesn't take over sound. I haven't actually explore it, but it looks more like what other solutions should have been, :) and perhaps easier for the sort of mixing at hand. I haven't ever gone beyond a single user on a single PC either without more complexity than having an usb extra sound device...