Giving vocals a consistent volume through the mix

[Babu]

Hello everybody! I'm currenly mixing and mastering an album by myself. I'm just an aficionado, so I'm working with my PC and Ableton Live 9.7.1. So far, everything is sounding great... Except for the vocals!
The performance is good and I have multiple takes. Furthermore, I've already edited everything to the grid. The problem is that the volume of the vocals is inconsistent: this may be either because the vocalist wasn't always in the same position/he moved the mic or because there was a change in the gain staging during the recording. Either way, most of the samples are in different volumes, and I'm looking to even them in order to have a more polished mix.
I've tried using limiters such as Waves' L2 + compression (Ableton's own compression) after EQing, but it still gives me a lot of trouble. What should I do? I thought of adjusting each vocal sample's gain, but it consumes a lot of time and it is not always an accurate method. I've also hear there's a Waves plugin called "Vocal Rider" that may help, but I don't know how to use it.

[Zukan]

Automation reigns supreme here but if you want something quick and easy then there are many plugins that can manage gains for you. I use Hornet's Auto Gain plugin that Matt recommended to me. Does the work for you and does it well. There are 2 more gain management plugins on his site that will match gains to a specified value or normalise to a target.

However, you can opt for the standard compressor techniques if it serves you better but I would, as stated, start with automation.

[Jack Ruston]

Automate, compress, automate again.

[Mixedup]

Automate, compress, automate again.

This. Plus maybe a couple of passes of editing and clip gain... and maybe another pass or two of automation

[The Elf]

Automate, compress, automate again.

I never felt the need to actually 'automate' (in the tradition sense) ahead of compression, but I do undertake quite a bit of basic level manipulation, and maybe mult out the sibilants and tame the pops before I go near any compression (and I'll often use multiple layers of compression).

I often also use parallel compression to give the vocals a solid foundation.

It may sound a bit OTT, but in actuality each stage of level control is about small amounts.

After that it's automation.

[Jack Ruston]

So, yes...when I say automate, compress, automate, what I mean is more along the lines of clip gain-type processing in the first instance. I'll roughly set a consistent level, first across sections, and then just to control anything that's far too loud or quiet within any given phrase. Then I'll compress...then automate the details as needed. The vocal always feeds a parallel compression bus, and then there's mix bus compression.

J

[Matt Houghton]

So, yes...when I say automate, compress, automate, what I mean is more along the lines of clip gain-type processing in the first instance. I'll roughly set a consistent level, first across sections, and then just to control anything that's far too loud or quiet within any given phrase. Then I'll compress...then automate the details as needed. The vocal always feeds a parallel compression bus, and then there's mix bus compression.

J

Well that's reassuring coming from a guy with your talent and rep... because that's pretty much how I tend to approach it too.

[Jack Ruston]

Well, great minds, Matt, great minds.

I will offer one more point...vocals don't want to actually be consistent. Most of the time they want to sound like they are, which isn't always the same thing, and sometimes they need to sound far more dynamic. So if you just technically even them out, you're going to have a problem UNLESS we're talking about a heavily distorted vocal, which can likely sit in one place. So why even it out? Because it wants to hit the compression in a flattering way, without bending over, and because often there are variances in the takes which need matching up a bit. Then we compress for sonics and energy. Then automate for a intelligibility, and to put back some of the natural surges in level.

[The Elf]

Sounds like a roaring agreement! Interesting that we've arrived at roughly the same solution along our different paths.

Just a quick note on parallel compression... I find it incredibly helpful for vocals, but most of the love I hear for it focusses on drums - for which I think it's far less appropriate. Just my personal take...

[Drew Stephenson]

I wonder if that's a reflection on the quality of drummers you normally work with?
It's not something I ended up* using with our band's drummer (he's very consistent - he hits everything really hard all the time!), but I have used it on my drumming just to smooth out some of the inconsistency between snare hits and toms in particular.

* Tried it and wondered why my drums were sounding a bit mushy and flabby...

[The Elf]

It is very much dependent on the attack and release settings you set on the compressor - you can add mush, or add punch (transients). It can be useful, and it can make the drums sound more trashy and/or exciting but it's not typically the way I'd do it.

I've used parallel compression to lift snare ruffs and rattles. Every technique has its place.

[Zukan]

Definitely a parallel kinda dude I be. Basses, vocals, drums and sometimes the whole mix.

I have this little trick whereby I bandpass the mix into three sections with each section being run off an auxiliary. I then treat the auxes to get the final balance and colour nailed. Works a treat and a really good way to master problematic mixes.

[Drew Stephenson]

And here we have another reason why I love this place, in a little side discussion off the end of the thread, we have all these handy little tips and techniques being dropped.

[Matt Houghton]

Sounds like a roaring agreement! Interesting that we've arrived at roughly the same solution along our different paths.

parallel compression... I find it incredibly helpful for vocals, but most of the love I hear for it focusses on drums - for which I think it's far less appropriate.

I use it frequently for both. On vox, partly it's kinda for what Jack said about not wanting it to be too level... the blend allows more consistency in the lower-level details, which can aid intelligibility internally within the part, or add an intimate whispery breathiness (or asthmatic wheezing in modern pop ). Then the last step of level automation is more about intelligibility in the context of other sounds that can mask details in the vocal... little nudges to keep the vocal on top (or to push down other things to the same end, sometimes). Ie the parallel compression is kinda giving me the sound I wish had been delivered and recorded, and the automation is making that sound work with the other stuff going on in the mix.

For drums, it's not about how good the drummer is. If a drummer is not hitting consistently the sound of each hit is different anyway, and I don't think teh compression can really even out such details and give you a tone that works. I'd rather find some nice hits and use them to replace the dodgy ones. It's more about injecting some energy/sustain/density in places, without squeezing the life out of the transients. Eg I might ride up the parallel compression in a heavy chorus and scale it bit back in a more open verse... it's one reason I kinda like having parallel comp on its own fader, rather than using a wet/dry blend control — I have the basic drum sound I want and then ride the parallel comp fader up when I need to add that to the mix, so it's adding energy, rather than just changing the balance.

[Drew Stephenson]

Thank you, I shall have a play with that technique too.

[CS70]

Definitely a parallel kinda dude I be. Basses, vocals, drums and sometimes the whole mix.

I have this little trick whereby I bandpass the mix into three sections with each section being run off an auxiliary. I then treat the auxes to get the final balance and colour nailed. Works a treat and a really good way to master problematic mixes.

Woah, never thought of that! Great idea to try!

[The Elf]

Somewhere I have a Cubase template that uses clever polarity trickery to negate the phase problems associated with creating a three-band filter split. If anyone is VERY interested I'll try to fish it out, but I will admit I can't recall how I came up with it - it works though!

[Hugh Robjohns]

Yes please!

[The Elf]

Yes please!

Bluff called!

Here goes:

[The Elf]

It's a few years since I created this, so I'm struggling to recall the details.

Essentially what it does is split the source into low, mid and high bands by use of filters and uses polarity inversion to remove crossover artefacts. It was based on a two-way mult someone showed me, but I gnawed at it until I came up with a three-way technique.

The 'original inversion' channel is simply to prove that no artefacts remain. When un-muted with all faders at unity the output shou;d be null - and it is!

[The Elf]

Have I successfully scared everyone into stupefaction?!

[Drew Stephenson]

We're cogitating...

Or waiting until we're at a screen where we can zoom in enough on the screen shots to see what's actually going on!

[The Elf]

[Wonks]

As clear as Brexit so far.

[Wonks]

Possibly because you've posted two pictures of a low pass filter, when you wanted one low pass, one band pass and one high-pass filter.