glcs icon indicating copy to clipboard operation
glcs copied to clipboard

Proper sound capturing solution

Open svenstaro opened this issue 10 years ago • 3 comments

The current suggestion to use some kind of alsa split is not a proper solution for most people. In order for this software to be properly adopted, this NEEDS to be simpler to get going. I'm adding this bug as a general enhancement.

This should be a priority, I think.

svenstaro avatar Apr 19 '14 01:04 svenstaro

svenstaro,

Since I know you, you have expressed this opinion several times. So I believe you when you say that you think that it is important!

I am going to repeat my position again.

If we put aside the ease of use for one moment, technically the alsa split solution is probably the best, IMO. Syncing and muxing several multimedia streams is a complex task and I prefer leaving that task to tools that are dedicated to solve that problem. Besides this is more in sync with the UNIX philosophy to which I adhere totally.

Someone has the right to prove that I am wrong with a pull request and I would be agreably surprised.

IMHO, a more fertile venue to address the issue that you express would be a tool that facilitate the ALSA split setup. I believe that this task can be automated and assisted by SW. It could be a script but this is just my personal taste. It could be a GUI program.

I created this issue specifically for that here: https://github.com/lano1106/glcs/issues/9

That would be great if you or someone else would like to look at that task but I will eventually look into it if noone else does eventually.

greetings, Olivier

lano1106 avatar Apr 19 '14 03:04 lano1106

Just to confirm, I see there is hook for asound / alsa in the source tree. This issue is about capturing all system sound right? If so I agree 100% with @lano1106.

Cloudef avatar Jan 19 '17 00:01 Cloudef

I haven't work on this project for a very long time. If I remember correctly. The main complaint was that seting up ALSA to create a split + loopback was too complex task for most users.

There is code to highjack ALSA API calls from the forked project but I stepped away from this as much as I could.

I choose to let the OS handle the sound. you create a split and a loopback that is fed to ffmpeg.

Yes this is quite complex setup task (but not that much) but it is the right approach. The way I see this could be improve is by automate and or assist the setup with some setup program.

lano1106 avatar Jan 19 '17 17:01 lano1106