multiple vocals take but always different levels

polkor

New member
Hi there
I try to figure out how to make two vocals sound on the same level.
I record vocal on two takes, fill-in the gaps but no matter what, there is always difference in levels.
I have cubase elements and izotope nectar but even though I apply vocal assistant still some parts are slightly louder other quieter (
need help. I even put tiny thread to keep myself in the same position in front of the mic.
heard about vocal rider but can't spend more money right now.
I am sure I am doing something wrong during recording because even software can't fix it.


disclaimer :dunno how to use automation :facepalm:

Screen-Shot-2020-07-27-at-11-11-00-PM.jpg
 
Not quite sure I understand what you are asking.

You don't need to automate though. Just use the scissor tool 'alt + left click' to separate the segments. Left click on the track to highlight. cut in between the vocal segments. That gives you separate events. Then pull up or down on the top center square to increase or lower the gain (level) of the event. Way quicker and easier than automation.

Hope that helps ya.
 
You can also draw in lines of gain in each event by right clicking on an event. Select the pencil 'Draw' tool and create points.

Have fun!
 
and don't forget that differences in level between takes is perfectly normal in a real singer - they're not robots. Are you sure you are hearing it, and not just seeing it?
 
You can also draw in lines of gain in each event by right clicking on an event. Select the pencil 'Draw' tool and create points.

Have fun!
Thanks for swift response. What i mean is that some parts even after compression seems bit louder than others. Sometimes vocal sounds like roller coasters thus i start questioning my recording techniques. I rap around 140 bpm so i need gaps to fill in vocals later otherwise i can suffocate during recording
 
and don't forget that differences in level between takes is perfectly normal in a real singer - they're not robots. Are you sure you are hearing it, and not just seeing it?
Thanks for your reply. Yes i can hear but surprisingly cannot hear in pro recordings ( other artist) hence I started to questioning my recordings techniques
 
Thanks for swift response. What i mean is that some parts even after compression seems bit louder than others. Sometimes vocal sounds like roller coasters thus i start questioning my recording techniques. I rap around 140 bpm so i need gaps to fill in vocals later otherwise i can suffocate during recording

You should never rely on compression to do anything. Get it right before.

That was my point man. Consider compression an effect. Not a leveler. Get that right first, or you will be using a compressor for all the wrong reasons.

Granted it can help with levels, but if you think of it that way only and don't understand what it is doing, it can really screw with what your goal is.
 
One thing you might try is to normalize both tracks to the same point. As long as you don't have some unusually large peaks in the files, it may balance them. If you have a large transient peak that will probably keep them from getting to a similar average volume.

You could also analyze them for short term LUFS and then boost the lower one until you get similar values. Use short term LUFS for short sections of similar vocals. If you are doing the same song, look at both versions of a phrase and adjust gain to match them.

If you're recording using the same settings and getting wildly different volumes, then you probably need to work on your technique. You are either varying your distances while recording or changing the "enthusiasm" of your vocal.
 
One thing you might try is to normalize both tracks to the same point. As long as you don't have some unusually large peaks in the files, it may balance them. If you have a large transient peak that will probably keep them from getting to a similar average volume.

You could also analyze them for short term LUFS and then boost the lower one until you get similar values. Use short term LUFS for short sections of similar vocals. If you are doing the same song, look at both versions of a phrase and adjust gain to match them.

If you're recording using the same settings and getting wildly different volumes, then you probably need to work on your technique. You are either varying your distances while recording or changing the "enthusiasm" of your vocal.

definitely you are right about enthusiasm :laughings:
i will try normalization ? frankly i have never used it because problem is between two separate vocal tracks
 
I've been chopping up opera in comping sessions, and the differences phrase by phrase in the takes are quite interesting. The practical thng if no electronic level changes with faders and gains happened is that the singer sang quieter or louder, or stood closer or further away. If tonally they're fine, then ride the fader!
 
definitely you are right about enthusiasm :laughings:
i will try normalization ? frankly i have never used it because problem is between two separate vocal tracks

Try normalizing each track to something like -3dB. That should make them a closer match in apparent volume. If its only a few phrases, it would be easier to just adjust the fader.

Automation isn't hard. I had a track from a few years back that had a really high tom hit on the overhead. I tried a few compressor settings but they didn't work well. There were only a few hits during the song, so the easiest way to fix it was to use automation and to put a big dip on those peaks. It took some time to go through the track, but in the end, it recovered a track that could never be replaced. It was either that or do without cymbals.
 
What everyone else said - normalize the tracks, or use "clip gain" to get levels consistent. Slice and create the "comp" with any necessary cross-fades. Use automation to apply any final smoothing.

THEN (and *only* then), create a single comp track (bounce to another track) or send the comp (sliced/cross-faded/automated) tracks to a single bus and put your FX on that. Don't do it to the individual tracks/takes.

Of course, if you use Logic or something that has a similar "take folder" concept, you can put the FX on the folder (automatically created comp).

P.S. This is one place I always remind myself to not move anything or take notes (which I am exceedingly poor at) because if you have to come back the next day or even later to punch in something, getting everything set up exactly the same will help immensely. Even if you think you have good takes, sometimes on the nth listen you'll hear something and go, ooh, I really have to fix that...
 
Last edited:
I really don't get this. If the song requires every phrase and syllable to be the same volume - shriek Metal, for instance, then normalise away, slap it through a compressor and turn it up - but those waveforms to me are close enough to match pretty well - but I suppose this is mixing by eye, not mixing by ear - the very modern way people do audio. They look and notice the 1 or 2dB differences and assume it's a fault.

Question - when you comp the two takes together and close your eyes, do you hear it go quiet/loud/quiet/loud?
 
Back
Top