Mixing a vocal harmony
Home > Home Recording Forum > Recording Techniques > Mixing a vocal harmony
Posted on Jul 08, 2009 08:40 am
Adam Mutschler
Abomb Muchbaby
Member Since: Jul 02, 2009
So I had my friend record a really nice vocal harmony to one of his songs. The timing is really good, the notes are dead on, but it seem like no matter what I do, I can't get them to sit well together. I have always admired what the chili peppers do with their backups and harmonies. so that is what I am shooting for I guess. If I make the harmony quieter, it doesn't seem to sound right, I have had some extreme settings and can't seem to find something that works well. Should I approach each vocal track individually? Or do i need to export them as a single track, and work it that way and then bring it back into the mix? Any thoughts. Thanks guys...
Adam
[ Back to Top ]
HuePinnipedal Czar (: 3= Member
Since: Apr 11, 2004
Jul 08, 2009 09:18 am This is what advice I could lend... others may do this completely differently, and I hope that they chime in as well.
Approach them individually, eqin'g-out what ferquencies arn't wanted in each track.(ie... roll of the bass, and notch-out any mud/noise/hiss) Allow two sends/sub-busses to each track... one for reverb, and the other for paralelle compression. Dial in your prefered reverb in one sub-bus, and send as much of that track to the reverb as wanted. Repeat, until you have each tracks reverb(w/o regard for the others) right where you like it. Turn down the fader on the reverb bus. Now, do the same with the paralelle compression bus. Turn it all the way down after you have each tracks added compression where you like it.(It should be just enough to add a little punch and sparkle.) Turn both sub-buses all the way down, and then work on getting your, now eq'd, tracks to to sit as well with eachother as possible. Start bringing up the coompression-bus until you just start to hear it work... do the same with the verb. Don't mix anything so hot that you kill your headroom. The verb will help things jive, along with the eq'ing, the compression will help even things out, level-wise.
Some things to look for that you can do/think of to help 'as you go'...
If there are any frequencies in one track that is stepping on anothers... give that freq a subtle notching-out... subtlely. : ) Also, the flavor of your reverb will have considerable effect on the mix, so choose well.
Hope this helps.
Jul 08, 2009 09:27 am Wow, thanks man. This sounds like a really good approach. I may have a question about this once i try and execute it, but I will let you know. Thanks a hell of a lot!
HuePinnipedal Czar (: 3= Member
Since: Apr 11, 2004
Jul 08, 2009 09:30 am Right on, Adam. Welcome to HRC, by the way!
CptTrippsCzar of Turd PolishMember
Since: Jun 20, 2006
Jul 08, 2009 11:13 am Hue layed out some great technique so you are in good hands. Another thing to consider, and what I do...
Sing the harmony/backup twice and pan them L/R leaving the main vox centered. If there are multiple harmonies, pan them a little further (or less) L/R than the previous.
EQ'ing them differently and a little panning are some key steps for seperation imo. This may be old news to you, but I figured I'd toss it out there.
Although I normally route all my vox to the same Verb, I do comp them seperately (mains, backups, harmonies) and love love love the parallel comp when applicable, is so smooth.
And yes, welcome to the HRC.
Jul 08, 2009 11:45 am thanks guys,
Not gonna lie, I am definitely new at all of this, so I am going to try and use this advice. Some of it seems that it might be a little over my head, but I am gonna give it a shot. And thanks, I am glad I found the HRC.
CptTrippsCzar of Turd PolishMember
Since: Jun 20, 2006
Jul 08, 2009 11:49 am What part are you thinking is over your head? The buss routing? parallel comp? Let me know which part is confusing, I have some time at work today and would try to explain the basics of a technique, that way you can research it more and not be overwhelmed.
Jul 08, 2009 12:11 pm Well, I use cubase (i think it's SX?) and I think I understand compression.
As far as EQ goes, I know that it's not something that can be explained in a few sentences. I think I understand the purpose and what it does, though I don't think I would say I understand how to properly apply it.
Sends and sub busses... Well, I don't really know what this is. I think different programs use different terminologies, I can put as many effects on a channel as I want, I think by sends it means just sending or having an effect on that channel?
Parralell compression... I think this just means using equal compression on each different track?
See when I mix something, basically what I do is start with the drums. For most of my recording, I will just use 2 overheads and a bass drum mic. This seems to give me decent results since I am not in a dead room anyway. So I just pan my two overheads left and right and run the bass drum right down the middle. Then I bring in the guitars. Then the bass and finally the vocals. I usually do vocals last, since I like to get the music to sound good by itself and then lay the vocals in on top of a good music track.
But when applying effects to anything, I do it on each individual track. Like each track has the option to add an effect. Is there a better way to do this?
I guess the part I don't get is the buss thing? or sub busses?
And as far as frequencies go, I am not sure that I know what I am listening for?
CptTrippsCzar of Turd PolishMember
Since: Jun 20, 2006
Jul 08, 2009 01:02 pm Okay,
EQ is your best friend when it comes to mixing and is how you seperate things out and give them there own space. This graph can you give you a general guideline of where things reside but playing around with it will be your best weapon.
Take bass and kick for example. Your kick may be prominent at 80Hz (low and boomy) but your bass guitar is going to occupy some of this same space. So you add an EQ to your bass guitar and simply cut a couple db at 80HZ so the two are not fighting for the same space.
This will apply to all tracks occupying similar space as another (two vocal tracks for instance).-
Once you get a general understanding and start to play around you will realize the power of EQ, it is one of the first things you should get a handle on to create a good mix imo.
I will try to touch on the others after a bit.
Jul 08, 2009 01:43 pm Damn this is cool. I once saw an illustration like this, but nowhere near as in depth. This is really cool though. Obviously you don't want to literally follow this everytime, i know that recording/mixing/mastering is as much of an artform as music itself... Thanks though man, Can't wait to try and take advantage of this.
TadpuiI am not a crook's headMember
Since: Mar 14, 2003
Jul 08, 2009 02:34 pm What I've done (with at least moderate success) is just compress the crap out of the backup vocals. This achieves a couple of things.
First, overcompression tends to rob some high frequencies out of the signal. This is good because the vocals with less high frequencies will sound farther removed from the listener, and it will make it so that the lead vocals sound like they're "on top" and not being buried or competing with the backup.
Secondly, the reduced dynamic range of overcompressed vocals make it much easier to manage them as far as gain/level goes. I find it easier to find a "sweet spot" of gain for the backup vocals when they're very consistent in their dynamic range. But this can also suck all the life out of the vocal part. That's why I only try this for backing vocals.
An old forum member here taught me that trick, and I've been using it ever since. Jues was the man, and he produced some very slick sounding songs.
You'll definitely want to use reverb as well, since that's the tool that you use to distance sounds from the listener. You want the backing vocals farther away from the listener than the lead vocal, so that the lead ends up sounding more dominant in the mix than the backups.
Even a little EQ will help. You can either apply the same vocal "notch" to the backups that you apply to all of the other other instruments in order to make way for the lead vocal, or you can do like was mentioned above and find different frequencies to attenuate and/or amplify for each vocal part so that every part gets its own slice of the sonic spectrum.
cooloFrisco's Most UnderratedMember
Since: Jan 28, 2003
Jul 08, 2009 04:10 pm I usually eq the backup vocal a little higher in frequency than the main vocal and take more frequencies out of it leaving it a bit thinner. Also put a little more verb on it so it sits behind the main vocal. And finally, some subtle panning to differentiate it just a bit more.
McMerkinHobbyist musician,pro recorderMember
Since: May 15, 2007
Jul 08, 2009 04:14 pm Agree with coolo. I have noticed that most b/up vox feature less bass frequency to them, also lending to the 'farther away' effect in conjunction with the added reverb.
Jul 09, 2009 12:30 am don't forget to pan them!!!!
keep yer main vocal dead center, then pan the backups out a bit for seperation....
HuePinnipedal Czar (: 3= Member
Since: Apr 11, 2004
Jul 09, 2009 08:18 am Quote:
don't forget to pan them!!!!
hahaha! Yeah... the 'panning' of your tracks. haha
That's pretty important, eh?
That will be very relative to what you're song tells you to do with them... I mean, how the back up vox enters into the piece, the levels of the backing vox, and such. What 'effect' the backing vox lends to the song will be drastically influenced by your panning of them.
Jul 09, 2009 02:08 pm Actually, I tend to double my vocal tracks and pan one left and one right even if there is no harmony. Is that bad? I know it all depends on style and taste, but is there a technical reason not to do this? I like the effect that it gives, but could it be taking away from the panning effect of the other parts in the mix? The only thing that I don't pan at all would be a bass guitar and a kick drum... I just always thought that panning the vocals hard left and right would give the mix a more wide sound... Then when I did my harmony I would pan them half left and half right...
TadpuiI am not a crook's headMember
Since: Mar 14, 2003
Jul 09, 2009 02:17 pm Double tracking vocals is a pretty common practice, especially with vocalists that aren't totally confident in their delivery. It makes slightly missed notes a little easier on the ears and eases the dissonance between the vocals and the backing music. Elliott Smith, even with his great singing voice, almost always doubled every vocal part he recorded. I've found the same to be true with some of my favorite bands, like Islands and Sparklehorse.
Some fun things to do with doubled vocals are to distort one of them, or severely compress or hard-limit one of them, EQ one to mimic a megaphone or telephone, or run one of them through some wild set of effect chains. You have the effected vocal for texture, and the uneffected one for clarity.
When doubling vocals, play around with the width of your panning. I find that the closer they are in the stereo field, the more dissonant it sounds when they don't both hit the exact same note. But when they get too wide, it gets a little distracting to the listener because the vocals should appear to be front and center in the mix. They're the focal point for the listener (most listeners, at least). So it's kind of a balancing act.
Now when you start putting in harmonies and backing vocals, your stereo field will start to get crowded in a hurry. Whatever you decide to do, just make sure that each voice has its own little spot in the stereo field, unless a "wall of vocals" is what you're after.
CptTrippsCzar of Turd PolishMember
Since: Jun 20, 2006
Jul 10, 2009 01:02 pm I double my main vocals as well but never pan them as lead vox are something I don't want doubles to be noticable. I want them to sound full but not like two vocals. Imo panning a lead vocal does exactly that. Instead of panning two lead vox, leave em centered and maybe try a small bit of delay to widen them.
Jul 10, 2009 02:12 pm Actually I should have clarified this. I don't actually record two tracks, what I do is just duplicate the track that I already have. Then I pan them. Occasionally I will sing two tracks, but typically It's just one that's essentially copied and pasted. But I will try keeping the vox centered, and I really like the idea of kind of making the two different for a cool effect. It's funny that you mentioned the delay too... I sometimes will manually delay it, meaning I don't put an effect on it I just back it up like a millisecond. Works pretty cool depending on the track...
TadpuiI am not a crook's headMember
Since: Mar 14, 2003
Jul 10, 2009 02:33 pm Any time you want 2 tracks of something, its generally better to actually record a 2nd take insted of just copying and pasting the same track. The reasoning behind this is that the subtle differences between 2 takes adds a depth and/or thickness to the part that you simply can't achieve with 2 carbon copies of the same take. The same goes for rhythm guitars (and especially for heavy rhythm guitars...double-tracking is the engineer's secret weapon for getting thick, heavy electric guitar sounds!).
Copying and pasting has its uses (mainly for the delay effect you're already experimenting with), but it doesn't achieve the same effect as 2 distinct takes being recorded and played in unison. Even if you nail both takes, there will still be subtle differences between the 2 takes that add a texture to it that's pretty much impossible to duplicate with copying and pasting.
CptTrippsCzar of Turd PolishMember
Since: Jun 20, 2006
Jul 10, 2009 04:51 pm Oooooohhhh....
Double track that sucker for real :) It is a much better sound I guarantee. If you don't get it right away, keep going. The copy paste thing is a bad habit you will want to get past.... now :)
What I like to do is sing the first one as a scratch. I do this so I can hear it when I do the second track, the fullness of it tricks my brain and makes me less self concious. I then go back and redo the first while listening to the second.
Keep them both centered and then add some delay if you want it wider/set back in the mix. The reason I don't pan doubled main vox, when they get too wide it becomes very easy to notice there are two tracks.
Let us know how things come out :)
