Issues with multiple audio tracks, what is the optimal way?

[bdbr]

Hello,

I am trying to record videos where I have a desktop microphone + computer audio. I would like to be able to just hit record, then the video created will automatically use the channel with both of those inputs. But in the background I'd like two extra channels, one with each input so if I got levels wrong during recording, I can then go and use the individual audio channels in a video editing program and adjust them accordingly. I'm encountering some difficulties though. If I do this:

Channel 1: Computer + Mic
Channel 2: Computer
Channel 3: Mic

After I record, the video will be using channel 2 or 3 (can't remember which one). How do I choose which of these channels comes through on the recording and isn't "under the hood"? I can't seem to find that setting anywhere in OBS. I can get around it by changing the channel that uses both inputs, but then I have confusion when converting the video afterwards.

Hope this makes sense

 

[dmemphis]

The trick there is to get a mix of the two items to show up as a strip in the Audio Mixer window.
I use Voiceemeter to mix sources and present them to OBS as a mixed output that is selectable
in a scene source.
I have heard of a new OBS plug-in mixer referred to as the "voiceemeter killer" that should do a similar thing. I haven't tried it.

Then you would bring the computer and mic into the the Audio Mixer by virtue of normal definition of scene sources or as global audio devices in Setup->Audio.

Now you have three strips in the Audio Mixer window.

The next trick is track assignment, which is done in the Edit->Advanced Audio Properties.
The subsequent recordings should show these as separate tracks in your editor.
That's the theory, I have not done track assignments.

Not quite a tutorial but I think that hits the important parts of the solution.

 

[koala]

In Edit->Advanced audio properties, you can assign your audio sources like this:

This way, you're creating track 1 as mixed track with all sources mixed, and tracks 2..5 with each source individually.
Media players will usually choose track 1 for playback if there is no preference, so you will be able to immediately hear your mix.
For postprocessing, import the other tracks in your postprocessing software and create the mix with this, then replace all audio tracks with your mixed track from postprocessing.

Keep in mind, no media player will mix audio on his own. Media players will only play one track at a time. To get a mix, you need to create one mixed track. The additional tracks are for postprocessing only, and in regular media, for providing additional languages.

 

[dmemphis]

Ah! RIght! no mixer needed in this case!

 

[bdbr]

This way, you're creating track 1 as mixed track with all sources mixed, and tracks 2..5 with each source individually.
Media players will usually choose track 1 for playback if there is no preference, so you will be able to immediately hear your mix.
For postprocessing, import the other tracks in your postprocessing software and create the mix with this, then replace all audio tracks with your mixed track from postprocessing.

Keep in mind, no media player will mix audio on his own. Media players will only play one track at a time. To get a mix, you need to create one mixed track. The additional tracks are for postprocessing only, and in regular media, for providing additional languages.

Thanks for this. I was struggling to get this working when using handbrake to convert, but I think I've realised I had set Handbrake to choose track 3 instead of track 1.