Export, stems and loudness (oh, my)

I have two types of Cubase projects to manage. There are the session projects for tracking and mixing, and then there are projects that just contain stems for the sake of rehearsal. An example of the stems might be drums, bass, keyboards, orchestra and backing vocals.

I have a Yamaha mixer in the live room that connects via USB into a laptop and shows up in Cubase as 32 i/o channels. This is for the stems projects, and the idea is that I can rehearse with backing tracks as well as covering situations where, for instance, I have a bass player but the drummer couldn’t make it. I just mute the stems for the guys who are there. So far, pretty simple stuff.

My challenge is exporting the stems from the Cubase with loudness that’s consistent from song to song. If I rehearse a dozen tunes, I don’t want people having to stop and adjust their IEM mix before each song to turn, say, the bass up or down because the levels aren’t consistent between songs.

I’d like to be able to go through each session project and export out the stems (which are typically represented by groups) in such a way that the loudness is consistent without having to introduce an additional step such as firing up Wavelab and doing batch processing on what I’ve exported.

Perhaps a two step process is what is required, but at version 9 Cubase is so feature rich that it seems like there must be a way to accomplish this in my exports. I’d be grateful for any advice you might have on the most efficient way to accomplish this goal.


You’d need to have preliminary stems of the group channels already present in the project as audio events. Then proceed as follows :

  1. Decide on your target integrated loudness value for each stem.
  2. Measure the current integrated loudness of each stem by selecting the audio event and then selecting Audio / Statistics. You’ll find the integrated value in the R128 section of the Statistics window.
  3. Calculate the difference between the current and target integrated values and then add or subtract this difference on the fader of each stem group channel.

Thanks, I appreciate the help.

That all makes sense to me and would keep me in Cubase. However, it looks like more work than setting up batch processing in Wavelab and doing a two step process. What I’m hoping for in a perfect world is an automated, “fire and forget” approach. I know I can do that by dropping files into a watch folder in Wavelabs, but I wasn’t sure if there was some way to get a similar effect when exporting in Cubase, e.g. “export at -23.0 lufs.”

No, there’s no export at -23LUFS. But you could just work to a standard when you’re in the process of producing your stems using the loudness meter in real-time. Most of the time you’d be able to get close to your chosen target.

But you’re right the whole thing is easier in Wavelab… (or Nuendo) if you want to hit a precise integrated value.

Yeah, it’s all about the right hammer for the nail in front of you.

As I think this through, when I export the stems from Cubase I guess I could just make the destination the watch folder for the Wavelab normalization batch processor. While it’s two programs and steps, in essence it’s still fire and forget, especially with the batch feature Cubase now has in the export window. Pick the stems, press export, then go get a cup of coffee while Wavelab does its thing.

Yes that sound like a good technique.

Well, you know, anything that involves more coffee.