Hi all,
I’ve been doing recordings lately for bands that have both vocals and instruments. After the show I export the mix with both instrumental and vocal versions and I’m wondering if anyone has a good workflow for this? Right now I’m mostly just exporting the mix, then muting the vocal mic and exporting again and performing volume normalization as part of the export. I’ve also tried having a second bus with the vocalists mic and all the instruments on it and swapping between that and the main bus when doing the export, which works okay and lets me normalize volume separately for each bus, but only the master bus allows you to use the loudness analysis tools as far as I can see. I also sometimes use compression on the bus with all the instruments sidechained to the vocals to make them a tiny bit clearer, so that ends up having to be turned off as well before exporting the instrumental version.
Is there a better workflow I should be using for this? I’d love to be able to more easily mix both versions of the song a bit differently and/or not have to run the export twice, but it’s unclear to me how to come up with a workflow that works well for this or if that would even be desirable, maybe it’s easier just to have two separate projects using the same source files and re-do the mixing in each.
Thanks!
I’m just curious why you’d be needing or wanting two separate versions (i.e. the instrumental version, and the instrumental + vocals version?) in the first place? If these are live shows(?), and vocals were performed, don’t you just want to mix and then export all of what was performed? O___o What am I missing here? 
Regardless:
You’re doing fine it sounds like. But work to set-up any side-chaining to effectively stop working as soon as the vocal track(s) is/are muted, so that bypassing plugins before export is a non-requirement.
Words can be confusing though… A screenshot (or two) of your editor might help to easily depict what it is you’re doing exactly, and thus what might be easier. 
-J
This is a type of folk music/dance with spoken word (or, more rarely, sung) vocals over the top. Some people just want the music, some want the music and the dance calls. Different people do it different ways, some CDs meant for practicing calling to come with the vocals in the left channel and the music as a mono track in the right channel, then you can pan to the right to get just the music (not ideal, but it’s how it’s done a lot, I prefer to just export two mixes and have 2 stereo versions of each track).
Yah, it technically works, it’s just a pain.
That would be helpful, but I’m not aware of any mechanism that would make that possible? Or maybe it already happens, I don’t actually know where in the chain the mute button sits but if the sidechain channel were after that I guess it would automatically do this and I wouldn’t have to worry about it. I’ll experiment when I’m next at my workstation, maybe just moving the sidechannel post-fader will fix this, that would certainly be nice to have one less thing to turn off!
I wasn’t able to figure out how to get a screenshot, Flameshot complains that some long service name is broken and others I tried (mate-panel-screenshot, gnome-screenshot, xfce-screenshot) show the screenshot but then freeze the moment you hit save and never save it. Unsure what that’s about.
To explain in more depth I normally record the session as one long track with separate caller and instrument mics. I normally add a send to the caller mic and then add a tiny bit of compression to whatever bus all the instruments go to (or to the instrument track if there is only a single instrument) with the microphone send into the sidechain channel. Then I put range markers for each track and in the export dialog I export all the ranges as separate files. I then mute the mic (and turn off the ACE Compressor on the instruments) and do the export again. This gets me both files, but isn’t super flexible (ie. I couldn’t have different mixing settings or normalization settings easily for the version with and without vocals) and is tedious.