How to make vocals sung at different volumes match

I have sung lines at different times, at different volumes unintentionally, but I want to have all of the verses and choruses sound more consistent; they all need to be the same volume.

Is this possible? Or do i have to go through all of the lines of the song and sync them up?

Well that is what Compressors are designed to do - among other things. However if there are big level changes between sections the Compressor will respond differently to each section so it is best get everything roughly the same before compressing. You can do that by using Normalize which will adjust the level based on the loudest sample. Or you could use Loudness Normalization which bases the adjustment on the audio’s Loudness Level. Most likely Loudness Normalization is a better fit for what you describe. If you haven’t really used Normalization you might want to create a test Project to explore how it works on different settings & audio. Don’t Normalize too hot, leave yourself some headroom.

A lot of folks would Normalize the Audio and then Render the entire Track so it’s all in one Audio file.

1 Like

You can also use clip gain to adjust the volume of the individual takes/clips. Even if you end up using a compressor, evening out clip gain will help the compressor do a better job.

1 Like

:disguised_face: :disguised_face:
NUENDO ALERT, this feature is not part of Cubase.
:disguised_face: :disguised_face:

But it is

:policeman: :policeman:
KNOWLEDGE ALERT - JM displays serious lack of feature knowledge.
:policeman: :policeman:

Sorry boys ‘n’ gals.

Sometimes its hard to keep up with all the fancy new stuff.

What I would do is ride the automation on the fader on that track.
(Or, more likely, listen to it in sections, and then manually enter normalization with the mouse/curve editor.)
That really shouldn’t take that long – maybe 30 minutes for a 3 minute track, tops. (A LUFS monitor helps, btw)

It also doesn’t need to be perfect, if you buss the clips to a vocal compressor. You could set one up with a long release, so it works more like an automatic gain control than a dynamics shaper, if you want.

What I find when I do takes in different situations, is that the different angle to the microphone will lead to different characteristics of the take, and that’s much harder to control for – Eq can get you some of the way there, but it’s not gonna be perfect.
Then again, I like autotune and ovox and harmonizer and … at that point, those differences are not audible :smiley:

1 Like

You’re right about position of the mic. I really should try to be more uniform in how far away from it I sing, at what angle, etc.

I will try normalization in just a sec. But first -

How do I do this??? I do not see Normalization anywhere as an option…

Look in:
Audio>Processes>Normalize

Oh! OK, now that I’ve got Normalize down, any chance you could explain what this means?

An Audio Event in Cubase always refers to a section of an underlying Audio file (that “section” can even be the entire Audio File). When you have multiple Audio Events in a Project each of those Events could be based on different Audio Files, or they could all be based on a single Audio File, or some combination of both.

When processing Audio, like with Normalizing, it is going to process the underlying Audio File. The message is telling you that multiple Audio Events are based on this specific Audio File. If you process this file it will impact all of these Audio Events which you may or may not want to happen. So it gives you the option to apply the processing to only this single Audio Event. The way it does this is to make a copy (aka New Version) of the Audio File which only that specific Event will reference and apply the processing to this new copy leaving the other Events unmodified.

Since you most likely want all of the vocals to be Normalized you can probably skip the “New Version” but it’s always worth pausing a moment to consider the decision & its implications.

1 Like

Can I come crawling back out of my hole and mention that Cubase Elements does not have Loudness Normalization? I know I said it is a Nuendo-only feature, which is wrong, but Elements only has peak normalization.
Still better than nothing.