Mixing Vocals - the part that never gets talked about

[ravisoni]

I was watching puremix.net's gearfest video on mixing vocals, and there were a couple things that I had never ever thought about. That may come down to my experience mixing things (next to none), but I thought for any home-based musician, such tips would make a world of difference.
Immediately the studio version of axis of awesome comes to mind (it just doesn't sound pro!) and the reason for that is not because they didn't get the technicalities right, I'm sure they did. But the "feel", for the lack of a better word, was missing. And isn't music first and foremost about feel?

Basically what Dupont (puremix.net) was doing, among other things, was reducing the sibilance and plosives, but also the hum on the "m" sound of the singer (because they had mic'd him too close) when he said "machine", and I thought well damn, that right there is the difference between pro and amateur. These subtleties. And boy, if you listen to the vid, does that make some difference!

So my question is: what techniques - and I don't mean reverb or EQ or compression - do you use to get your vocal sounding better, whatever better means to you? I guess this would be the question about what you do to your dry vocals before you start sweetening it with your routine chain. Do you like your dynamics? How do you flatten your peaks? Or do you?

It'd be really neat to know, though I understand that your techniques make your own sound your own and you may be hesitant in sharing it.

[Some Desperate Glory]

I high-pass the hell out of my vocals so they even sound a bit thin when solo'd but that way they fit the mix much much better.  I got rid of that close, amateur sound that my vocals had.  That was the best thing I ever started doing.  I know you said not EQ but it makes a big big difference.

I also got a decent mic.  Not even that expensive.  It's a Rode NT-1 and it made an enormous difference on the quality of my vocals and it was like $200 or something.  Well worth it.

[Benedict]

1. Good Performance (can't ever fix that in the mix)
2. Good Balance to the rest of the mix (nice and clear so the story can be heard easily)
3. Good Mic

[ravisoni]

I guess I didn't explain myself clearly.
I'm talking about the other subtle things you do to enhance the vocals. Good performance, mics and overall mix balance are sort of foregone conclusions.
I'm talking about the... little things? Like take the hum out of nasal sounds, keep the sibilance instead of getting rid of it, etc.

[normen]

I guess I didn't explain myself clearly.
I'm talking about the other subtle things you do to enhance the vocals. Good performance, mics and overall mix balance are sort of foregone conclusions.
I'm talking about the... little things? Like take the hum out of nasal sounds, keep the sibilance instead of getting rid of it, etc.

The difference is that he heard what he wanted to do and what the problem was in this case and acted accordingly. Which is also exactly why you can't give tips like "boost 2kHz for a better vocal sound" because that won't work on every vocal.

[Benedict]

I didn't miss your point 

I only ever fix problems when I encounter them. I don't go looking as that to me is getting off course. If it doesn't strike me as wrong then it isn't and doesn't need fixing.

A common thing with poor initial performances (and micing) is to have to try to restore presence in the mix so that is commonly a combination of Compression and EQ (or even Saturation/Drive) which again I will assess and experiment with in the situation.

My point is that I see people thinking that the little tricks are the secret when in fact they aren't. Thinking that there are tricks leads people to looking for problems so they feel they are making things better by being active and involved, when getting an active and involved performance would mean that "problems" weren't an issue.

Selig recently talked about how the super-loud breaths in Steve Miller's "Jet Airliner" could be a "problem" or part of the performance. I couldn't image the song without them. Till he raised them I had noticed but never even considered them any sort of problem. If they were an error then they were best left in.

[ravisoni]

Thanks again for the reply Benedict.
I understand and in fact believe that learning the tips and tricks is not the only way to improving mixing skills. Again, my question is not as much about the tips as it is about what you, other mixing engineers, encountered.
Often times you will be mixing something (or seeing someone else do something) that sticks out to you as different.
Knowing that and keeping that in mind often helps troubleshoot the amateur imo.

For instance, I once wanted a snare like the one heard on Enigma's Same Parents. At that time I didn't know enough about distortion (or even EQing), and I talked about it with someone. They walked me through their snare processing chain, and told me, "you should really be able to tell yourself what it is that you're after, and know your tools." And to me, that was brilliant. Being able to talk it out, and give it a "name". I remember these little advices, and they are really useful to me when I'm stuck mixing a song.

I believe you gave me some feedback on a song I posted a while back, and I can't really remember what, but you said something, I tried it, and found it is applicable in quite a few situations. And that's what I was - still am - looking for, something I can keep at the back of my head if I get stuck again.

Normen: apologies if my tone came across badly.

[normen]

Normen: apologies if my tone came across badly.

 Uh, I meant what I said that way and didn't mean to "put you in your place" or anything?

Anyway one of the more interesting things I had was when I had a singer who would have funny sudden "bursts" of low frequency content in his voice. I think that was one of the only times I employed a multiband compressor on a single vocal track simly because pulling out those frequencies took away too much of the performance and simply compressing the vocals messed too much with the mid frequency range when these "bursts" occurred. It wasn't really a resonance type of thing, it pretty much moved around all over the sub-400hz range.

[MSD]

Vocals are a tricky subject, what you do will be dictated by the vocal you are working with AND the music that goes along with it. If it is your own vocals you are talking about then invest some time into getting it right at source, a good recording will do a lot of the work for you and a bad one will have you chasing your tail and probably settling for something that doesn't make you smile. 
The things that make all the difference will vary from track to track but there are things that you will probably do, to a greater or lesser degree, to most vocals that you are presented with.
Things such as:

• Low pass filtering. Don't cut too low. Usually between 100hz - 140Hz is a good place to look, if you feel you need to lose more, try EQ cuts / notches instead as you may lose important stuff if you filter too aggressively.
• Compression. Don't be afraid to stack compressors in series, sometimes two doing less is better than one doing more. Plus you can add the characteristics of more than one comp to your vocals, which sometimes can be as nice as the compression itself.
• EQ - There's no easy answers here, read articles and watch videos. There are a ton of problems you may need to fix and no two vocals are alike. Learn about presence, diction, warmth, body, nasal, sibilance, plosives etc etc. If you need to add a lot more of something, consider using a parallel channel. 
• Sibilance. There are lots of ways to de-ess, there's a factory soundbank combi, there's re's such as seligs and don't rule out automation. It might take a little work but drawing in automaton to lower the volume every time an S hits might work better in some cases. You can even combine a de-esser and volume automation.
• Reverbs and Delays. They are both great tools that can make or break your vocal mixing. Both are a very big topic but some things you can try; sending delays to a reverb, automating your delays to only effect some words / phrases, EQ your reverb / delay returns, some like to use the same reverb on the vox that they have used on the snare and resist the urge to drown your vocals in too much of either. 
• Listening. The best thing you can do to a vocal is listen to it, critically. This takes some learning but is essential to good vocal mixing. Listen for the things that are right and the things that are wrong, learn how to fix the latter while enhancing the former.

 
I know the above wasn't really what you were asking for but there really are no single techniques that will transform a vocal into something amazing. Learning and practice will yield results. 
Just to show that I am not a complete tool, here's some info on how you might set up a vocal exciter in Reason 

[djfm1983]

• Low pass filtering. Don't cut too low. Usually between 100hz - 140Hz is a good place to look, if you feel you need to lose more, try EQ cuts / notches instead as you may lose important stuff if you filter too aggressively.
You mean high pass filtering not low pass filtering. If you set a low pass filter at 100-140hz you wouldn't be able to understand any of the words. High pass filtering at 100hz (for example) would cut everything below 100hz ( depending on the HPF slope) and let all freqs higher than 100hz pass through.

[VNUprod]

Hey rs,
for me personally, there are two techniques, that work wonders and are often underestimated by less experianced mixers. I use those excessively, yet differently on every mix I do. And they can be useful on almost everything, not only vocals.
1 Automation: Volume balance, damp sibilance and plosives, automate certain EQ for a certain area of the track, panning, reverb
2 Parallel Processing: EQ, Compression, Distortion, Saturation, Autotune, Reverb

[ravisoni]

MSD, thank you for the lovely input!
All of what I was asking basically boiled down to your last point: what is it that I should "listen" for? I think this is how I should've phrased the original question.
To me, a podcast sounds fine, and sometimes even a telephone convo sounds fine. But obviously neither are acceptable form of a full-fledged vocal track in music, so what is it that straight away sticks out as something that should've been taken care of? Even the casual listener is often able to tell that something is "not right".
Being unable to quantify it was what was frustrating, but you've provided quite a nice list of things to watch out for!

VNUprod - parallel processing for vocals is something I haven't tried; perhaps it'll be something good to try out next time. Thanks!

[normen]

Well if you're the mixer you'll have to develop some idea or taste regarding that matter.

I had a real eye opener in these terms many moons ago. When I studied one teacher was switching between two EQ settings for a guitar track and asked us "So, what do you think is the better setting?" the answer was basically "Well we don't know." then he said "Then who do you think is supposed to know? You'll be the engineer responsible for the mixing - you'll have to make your mind up about such things and have a decided opinion about why you think what you did is the better option, how else are you going to sell your work to anyone?"

[MSD]

Low pass filtering. Don't cut too low. Usually between 100hz - 140Hz is a good place to look, if you feel you need to lose more, try EQ cuts / notches instead as you may lose important stuff if you filter too aggressively.

You mean high pass filtering not low pass filtering. If you set a low pass filter at 100-140hz you wouldn't be able to understand any of the words. High pass filtering at 100hz (for example) would cut everything below 100hz ( depending on the HPF slope) and let all freqs higher than 100hz pass through.

Correct, inattention to what I was typing is all. Thanks for the correction.