Carving the Mix for the Vox (Artists and Producers)

well, I try to make it sound good, and dial it in naturally, and use the country preset in Ozone.... oh wait, you said DON'T tell you that.....

Seriously..... I do not obsess over the tracks in my tunes whether they are instruments or tracks. Aside from getting the music accurate... no wrong notes, try not to use scooped to the pitch vocals....you know...that sort of thing....I really just record the vocals with a flat EQ setting and dry, with no FX applied to the incoming audio.

After the fact, in the mixing phase, I generally add Studioverb 2 dark plate @ about 20% setting.... that is a fairly light verb.

I also add Ozone4 to the FX bin...and yeah...I have taken a preset, done my tweaks to  it and saved it as "My Country Vocal" preset. Tweaks to the EQ, compression and a few other things. I almost always tweek that preset more to get it exactly like I need in the song. No 2 songs require the same exact settings.  I always need to tweeze it one way or the other.

I could open it and give you the specific numbers but in reality, you know enough to know that would be useless to you. Your voice and mine are different and you'd be just as well off to start from scratch tweaking a preset for vocals as I would.

Since no one really seems to like the sound of their voice, perhaps it would be better to send a dry wave track of your voice and a track of the song you wish to sing to someone you trust, who also has the same mastering tools you do.  Let them pop the 2 tracks into a project in their DAW. Allow them to work on the vox with the same tools you have.... and then when they send it back, you can see exactly what they did to the vocal track's treatment.  they can also save the mastering into a preset and send that in a file as well so you can examine it in your own DAW.

I like to get the music sounding good and often separate busses are applied in that process. But instead of sending the separate busses to the master I might create another buss and call it backing if you like and route that to the master.

I like to get vocals sounding pretty nice in solo mode. Do what you have to do to achieve that. I firstly edit vocal takes in a program like Adobe Autition. Just even a few things out level wise. Then in the mixdown channel some limiting first but with a threshold set high to only catch loud peaks. EQ to get the tone I am after and then some compression maybe. And of course time based effects can be applied. Any nasties in the vocal sound must be sorted out here.

I mix vocals in with the backing buss at low volume on a small speaker and go for that minimum voltage, maximum illusion approach. VU's don't change when the vocals are fully in.

Having all the backing or music on its own buss allows you to maybe insert an EQ with a light notch around the vocal area if the music is punching a little hard mid range wise, or you can tone things down on their individual busses eg guitars etc..Use you ears to tell you what instruments are interfering the most if you feel things are interfering with the vocals.

Editing what the music is up to as well while the vocal lines are present and when they are not also can make a huge difference. Even automating music or instrument eq changes can help a lot to allow a little more room for vocals when they are present and then eq change up in between etc..Use the vocals as a sidechain to duck a compressor slightly on the backing buss for example or one of the more mid range busses.

You should be able to keep the music sounding full range and well balanced and also if you have got a nice vocal sound then it does not need to be that loud in the mix before it becomes very clear and audible.

Thanks Philip for your thoughts. Yes my signature is about a great book that I have read twice now and still counting. Its called 'The Power of the Subconscious Mind' by Joseph Murphy. It is a great book. So much of the world that exists in your outer experience is totally dependant on your thinking. Change your thinking and you change your destiny. It is so true. I have recorded all the prayers in this book and the summaries at the end of each chapter. Once you get into regular subconscious mind work your world changes. You have to put time into working with and training your subconscious mind. If you don't, you are at the mercy of your negative thoughts then and things don't work out so well. And no plan means you could be going around in circles. But anyway....yes very interested in this stuff...

Maximum illusion minimum voltage refers to the concept of when a very important (lead or melodic especially) part of the music comes in and becomes very audible, the average voltage on the stereo buss does or should not change much. I like having real VU's tracking my main stereo bus all the time. They show a lot. If you have got your rhythm section happening nicely and it is mixed well the VU's will be moving a certain way to the music. Mixing in vocals through a small mono speaker at low volume means they become audible very quickly but the rms or average level on the stereo buss is not changing much. A synth pad is another good example. Very low in the mix and you will hear it quick but the VU's are not going to show it much if at all. If you have the vocals too loud they too start to wrestle the VU's and you will have problems with the vocals clashing with the backing more. They are both fighting the VU's. Let one aspect of the music do the VU stuff, (rhythm section) and the vocals do the illusion part and leave the VU's alone. (Another story for all the more reason to have real VU's over your mix. Peak reading meters found in many DAW's although important, fail to show any of this)

The powerful instruments like drums and bass will be contributing to the overall VU movement so we have a maximum illusion, maximum voltage situation. After that though more instruments represent the maximum illusion minimum voltage mode.

I find sending tracks to stems is a good way to make some overall changes to a particular stem. It only takes a very small dip of - 1db over a low Q wide band dip centerd around 1 KHz to completely change the sound of stem and suddenly push some things back nicely to allow for voals to come through easier.

Guitarhacker


Since no one really seems to like the sound of their voice, perhaps it would be better to send a dry wave track of your voice and a track of the song you wish to sing to someone you trust, who also has the same mastering tools you do.  Let them pop the 2 tracks into a project in their DAW. Allow them to work on the vox with the same tools you have.... and then when they send it back, you can see exactly what they did to the vocal track's treatment.  they can also save the mastering into a preset and send that in a file as well so you can examine it in your own DAW.

If you haven't you should take this route - if at all possible...

Hi Philip,

I know this is an older post, but I think the techniques part of the Sonar forum is where I'm going to hang out from now on. I'll be more help to people here anyway as this is what I do. :) I'd like to answer some of your questions in the manner in which I have dealt with them and experienced them myself. This is probably going to be a super long post, but I promise you a really good, informative read. :)

"My vocs seem kind of shreaky on the ears ... (noching the EQ at 3.5kHz doesn't always help)"
 
Vocal shreak can come from quite a few things. It can be that particular mic you are using, the angle in which you deliver your voice, how rough sounding your actual voice is, your actual vocal delivery (meaning, do you push like Michael Bolton to where your esophagus is going to leap out lol or do you use your diaphragm.) and of course, the eq curve you use on your voice.

For example, I have a U-87 Gold here. That's a pretty killer mic. But for me, it sounds horrible on my voice unless I sing in the key of G. Can you believe that?! I can't either. Something in that mic brings out a bit of a raspiness in me unless I sing in G. I don't know how it's possible or why, but it just is. So that said, the mic you choose can be VERY important. It's always a good idea to have a few good ones around. These days my go to mic for myself is an Equitek CAD E-200. It sold for about $500 when dinosaurs walked the earth years ago, but something about that mic just compliments my voice and I love the sound of it.

Other forms of "shreak" can come from not using a de-esser. This is important and can really clean up some of the shreak you mention. Your distance away from the mic and angle are important too. Sometimes the sound of our room can wreak havoc in a vocal presentation. I like to be close to my mic. The pop screen is 3-4 inches away and I am about 2 inches from the pop screen when I sing. This gives me an up close and personal sound without my room getting in the way. Now there are times where you may want some room to creep into the picture. You have to evaluate how good that room sound is. To me, I'd rather have control over that room and to do that, a good impulse sorts me right out. Or, I use the UAD EMT 250 which is a fantastic sounding verb. The key to using verbs and impulses is to create room atmosphere, not a hall or a tunnel. Long decay times are a thing of the past. We want "natural" to the point of no one saying "dude went over-board with a reverb on that".
 
"Please don't respond with the fallacies, "If it sounds good it is good" or "I just dial it in intuitively", "I use such and such a preset and get by".  Instead, try to state your strategies for your genre, or such."
 
Unfortunately, some of that fallacy is actually truth. Dialing things in intuitively really is the key and here's why. If you solo up an instrument and eq it that way, you are eqing that instrument to your ears...you are not eqing it to fit the mix, understand? The key is to make things work intuitively with the mix. So you want to eq with everything rolling at once. The only time you should solo an instrument is when it is a problem area. You know, something that is so wrong it should be re-recorded, yet you try to salvage it so it needs work? Those are the one's you can solo up because the other instrumentation may be masking the problem.   
 
"Do you consistently EQ and compress vocs on the busses or tracks?"
 
Yes sir, but never compression on a bus. I'll tell you why. To me, compression (unless you are sending a group of vocals to a bus or are using parallel compression) is an effect that should not have a mix of wet and dry signals. Not for processing of this nature. Vocal compression as well as vocal automation are what brings a vocal to life...and of course, the right eq and effects usage. But compression is the most important tool in the box for a vocal track even more so than an eq. I say that because I will not record a vocal that does not sound good or stand on its own. The mic I choose compliments the voice. Why would I record something that needs to be tweaked for hours? See my point? You NEVER record something and tell yourself later "I'll fix it in the mix". You waste so much time here, it's not even worth it. 

For example, I just recorded some scratch vocal tracks for a new CHB song. The vocal take was so good (the sound not my performance...I'm not a very good singer lol) it has 0 eq on it. Sure, for a final mix, I'd definitely tweak it some. But all it has is compression, a de-esser and a slight delay. The vocal by itself without anything but the compressor, stands on it's own. It still doesn't sound good due to the singer singing it..lol..but the quality of it as well as the natural delivery of the track, sounds good.

Now compression on busses to me, is a no no for a lead vocal. This needs to be applied as an entity. We want that compressor to work with the vocals. When we bus a compressor, the compressor and the vocals hold hands. When we apply a compressor directly to an effects bin, the compressor and the vocals become one...and entity so to speak and that's what we need. Sure, there are ways you can use it in a bus, but to me, it is way more consistent when handling vocal transients to be directly on the track. Once it handles the transients to where it's nice and consistent without over-comping, you automate to fine tune it. As for eq, you don't want excessive low end...you don't want loads of mids, and you don't want highs that make it hissy sounding. What to fix or adjust on this, depends on the vocal presentation...it truly does. I'd be shooting in the dark without hearing one of your vocal tracks to tell you where to fix things.
 
"Do you multicomp, say at 3.5kHz, to downward compress the shrills/sheaks?"

I would never put a multi-comp on a vocal track. If I did, it would be because that vocal track is in need of surgery and should be re-recorded. Multi-comps to me are surgical tools as well as mastering devices. There should be no need to run them on a vocal track in my opinion. Shrills and shreaks need to be identified before the vocal is printed. 9 out of 10 times, the mic you select can curb this artifact...honest.

As far as where to cut or boost, anyone telling you actual frequencies is going to be out of their tree. Each voice, mic, vocal delivery, mic pre and capture, is going to be different. You don't just jump on 3.5k if that frequency is not a problem area. You have to sweep through your bands and find out where the offender is, then make the appropriate correction. That artifact could be in the 7k range, why touch 3.5 if that's not where the offender is, see my point?
 
"Do you notch your EQ in the drum-buss or instrument buss (e.g., at 2kHz?) to allow clarity in 'some of your mixes'?"
 
I rarely touch the eq on my drum buss because all the eqing has been done on each track. On my drum bus is a Fairchild compressor to just keep them tight. It's not a parallel compression trick...it's just there to keep the transients in tact. However, there is nothing wrong with doing a little eq work to your drum bus if you feel the need. I work so hard on my individual drum tracks, I never feel the need for this. I do eq like crazy all over my drums though. High passing and low passing all over the place. If an instrument like say a kick drum, doesn't need 10k and above in it, why have it there? Removing the excess stuff you don't need in a sound makes your mix cleaner and makes room for other things. We don't need sub low end in a snare drum...why have it? We don't need high end in a bass guitar past 6k *most* times...why keep it? Most of the clack you want in a bass guitar is going to be from 1.2k to about 3k anyway. In the rare example, you may need a little something in the 4-5 k range just for some air or presence depending on the bass.

All the instruments walk hand in hand and need to be carved so that they all work well with each other. This is how you can hear everything so clear in professional mixes. They remove all the stuff that shouldn't be there...they give each instrument its own pan field, they eq effects, they control how wide an effect will spread, (this is important. Did you know that everytime you add a stereo effect that it is equal to hard left/hard right? Add enough of them, and you are walking on top) they don't boost frequencies in like-instruments and they make decisions as to what the instrumental blue-print will be BEFORE they even record most times.

For example, I had a great conversation with Philz last night on the phone about kick drums. When you are working up the core of your rhythm section, it is VERY important to decide on things before you start to mix and produce your material. Kick drums and bass guitars are easily 2 of the greatest offenders in the recording field. For 1, because they both are low end based and for 2, because they both usually share a center pan field.

The way to stop frequency masking is to decide what these instruments will be doing. For example, if we want a thuddy kick drum with nice low end, we shouldn't go for a bass guitar with that oooooom type sounding low end. It would be better for the bass guitar to have a little less lows, and a bit more clack and you let the kick drum handle the thud. If you decide you DO want a bit more ooooom sound to the bass, you create a kick drum with a little less thud and a bit more beater attack. When you do this, you make sure that you are not accentuating the same low end frequencies, compress accordingly, and you should be golden.

"How many layers?"

Unfortunately, you'll have to get back to me on what that question means as I don't have a clue. LOL! The only time I've dealt with layers, it was in regard to several guitar layers, back up vocal layers etc. So fill me in there and I'll try to answer it for you.

Sorry for the novel man, but hopefully this will give you a good idea on what you can do with some of this stuff. If you have any problems or questions, please don't hesitate to ask. :)
 

You're quite welcome, Mathew. Yeah it is a lot to think about. But once you get this stuff down and have experimented a bit, the results speak for themselves and you can really enjoy this field so much more. I think one of the worst feelings for me (other than being rejected by a hott chick lol) is years ago when I didn't know this stuff, how much time I'd waste and still not be happy after weeks of mixing one freakin' song! Ever have that happen?! Man, it's the worst! Even when you think you did a really good job and feel confident, your palms are sweaty as you put that CD in your car...you cross your fingers, you hope that you don't have to write anything down on that note pad you brought with you to take back into your studio and make changes on things that you can't hear in there anyway....lmao! I soo don't miss those days.

The next step in really fixing issues is to identify musical frequencies as syllables in the english language. This is how I teach my students how to listen to sounds. Each eq frequency is a syllable. It makes a sound and the more you mess with them and get familiar with them, the more you literally start to talk in "freakwency" as I call it. LOL! I don't know what instrument you play if any, but one of the cool things for me was when I finally knew every note on every string of my guitar. From there, my challenge was to memorize the sounds of chords and burn them into my head. I knew I'd never have perfect pitch, but I'm darned close and the things I've memorized have helped me. Just like memorizing frequencies. I'm never spot on with them, but darned close. I'll hear something and say "well, there's definitely too much low mid congestion around the 200Hz range." When I check the file, it may be 217Hz or 230Hz. I'm always close.

But my point is, it's good to know this stuff because once you can learn to identify and know what to listen for, it (sadly sometimes) makes this way too easy because the obvious is really right there in front of us. :)

Philip


Danny (and all),

By layers (vocal), I meant vocal tracks for the lead vox.  For example, I'm mixing a lead vox (Clifford) sent to me in Sonar 8.5 ... that Clifford has dutifully premixed into 6 vox tracks (of himself chanting):

1) Lead                   (centered)
2) Lead bkging 1    (centered)
3) Lead bkging 2    (centered or nearly centered)
4) Lead screw'd    (centered)
5) Lead LT            (LT 90%)
6) Lead Rt            (RT 90%)

Of these 6 tracks ('layers'), I detect that about 3 of them are seperate takes (depending on the section)

He EQ'd/Compressed them on 4 different busses (with very little track fx's) ... I'm quite satisfied with his 6-layered vox and the method he used.  But, I'd have preferred, as you, EQing and Compressing on the track level only (if CPU is not strangled)

Normally, for the lead, I work with 3 tracks (just 2-3 takes) (one having a 32msec Haas delay) ... 1 centered and 2 panned LT &/or RT. 

I realize a lot of intuition does occur for seasoned mixing artists, but novices and intermediates and advancers stll seem to require presets, paradigms, and/or "core strategies" (as you necessarily outlined with Philz for CHB).  Everything collapses on you, the mixing artist.

Your last post, btw, IIRC, reminds me of Mark Baxter's venerable paradigms ... who I think of as "Old School" ... but cherish his naturalistic diaphram-larnyx remedies -- haha!  (Heck, I think I've attained 2 octaves up, thanks to his survival guide ... haha!)

Applying "New School" techniques, I don't mind making sweet vocs out of inferior samples, as long as there are strong performance emotives (like the Talent gives a rat's behind).  Rarely have I had pitch-performance issues, timing performance issues, and/or dynamics performance issues with today's technology in the hands of a loving and patient mixer (that might be you and I). 

I'm in the camp that 'knows/believes' that the passionate and patient mixer is the real vox ("luv his chops" -- haha!) ... despite the samples he re-gurgitates (his own vox or others).  IOWs, you the composer-producer, with all your knowledge and intuition are the spokes-piece, the alpha-omega (haha!), the painter, the ELO-capitan, etc.  Thats why having a "core" groove-chord progression will sustain and flatter almost any *devout* vox, in my mind. 

Like recording real drum performances with special rooms and mics, vocs seem to demand abstracting into layers for vox-remixing (as a (co-) dominant element) ... which I'm learning.  (I state (co-) dominant element because producer-loop-grooves are pretty dominant in most mixes.

Danny, you've stated many excellent pearls and proverbs worth reading and re-reading:  Mics, Reverb pearls, delay pearls, compression, etc. ... which require repetition for us aspiring and seasoned artists ... who easily forget the intense passion, patience, artistry, techniques, etc. required.  At some point I'll ask you to crit (publicly or privately as you prefer).

 
Ok Philip, this sheds more light on the subject now....excellent! :) Let's see what we can do. I'll try to keep this post short...but you know how that ends up. LOL! One of the most important things to me in this field is to properly create your sound stage. This part comes in right after I choose my instrument blue-print most times. Picture a stage as you would see a show. Create this in your mind or map it out on paper if need be with a picture of instruments on a stage or draw something up really quick. Keep in mind there are no rules other than your creativity and everything I'll share here will be my personal, subjective views. :)
 
Pans are so important and the key is not to exhaust our wide pans unless we have to. For 1, the wider we go, the more separated our mix becomes. Ever hear drum kits with huge panning toms, rides hard left, hats hard right etc? This makes the drum kit become separated from the mix....too separated for my liking. On a real stage, our drums are centered. Though we will hear toms panning, hats on the right, ride on the left (listener perspective) it's not going to be this drastic pan we sometimes hear in music. If we keep our drum pans tighter, we can tighten the double rhythm guitar tracks that we created that most people run at 100L/100R which literally makes them smaller sounding. Pan a guitar all the way to the left. Then pan it 75 left....the sound is no longer so small.
 
Hard pans in my opinion should be left for special effect type things or for when you need something to be on the outside of your sound stage. Those are the pan fields that will really leap out at a listener. If we use them too often, the mix pretty much stays one dimensional and can bring on ear fatigue. There is really not much you can do for wow factor with too many hard pans other than bringing something in with a stereo enhancer that simulates a space that isn't really there. In turn, if you're not careful, this can walk over top of something.
 
Now the same can be said when you have too many center pans. What happens here is, your what we call "phantom center" is accentuated and at times, you can get a "phased" type sound going on. Let's take a look at your vocal tracks for a minute so I can further explain the method to my madness here. :)
 
Now we have 4 lead vocals centered in this. Along with this pan field, we have a kick drum and a bass guitar most likely. This makes the center pan field loaded with action. It's not a bad thing, but if you're not careful, you could have some things masking others depending on your eq curve. For me personally, I have always been a fan of "my voice is my voice". I never double track my lead vocals because if you are not super precise, it gives you that Ozzy doubling type effect and I sooo hate that. It also can form a dissonance within the vocals if extra care isn't given. The HAAS thing you mention is fine, but you also want to control the amount of spread on that because you really don't want your voice to take up the entire stereo image. Let's look at how to control that a bit more.
 
As I had mentioned in another post, any effect you run in stereo is going to be the equal to 100L/100R. Too much of this is like putting instruments all in the same pan field. So to fix this, do your HAAS vocals the way you always do. But before we go on, do you notice when you HAAS them, one side always seems louder than the other? The reason for this is due to the delay on the other side hitting later than the other. So it gives you the effect that one side is louder than the other. Especially at 32ms, you will really notice this volume difference and it could also be a little too loose. I like to use HAAS effects at about 24ms because it makes them tighter...but this is subjective. But getting back....to control the spread of these vocals, put a Sonitus Phase plug at the end of the chain. You don't have to mess with anything other than the width control. Remember that the width control here set at 100% will leave things unchanged and at 100L/100R. So, take the width down and you will notice the vocal spread will get tighter. Try to make those lead vocals sit right in between your hats and ride on the drum kit. That's the perfect space for a vocal to be in. You should be able to achieve this with a spread on the phase plug using 40-60 on the width control depending on how you have your drum kit panned. This lets the lead vocal sit nicely in the stereo field and it will actually sound a bit thicker because you are bringing the HAAS effect in tighter. The more delay and spread you have, the wider the vocal, but the looser it will sound as well as be too separated.
 
For your backing vocals, unless you have a choir of angels singing, I like to keep them on the outside of the hats and ride and in between rhythm guitars. This way they never walk on top of the lead vocals and they have their own identity in the stereo field. This leaves room for guitars slightly wider than the back up vocals, and if you have keys...you can have them wider to add some nice ambience to the mix. Quite a few people like to hard pan their boards...you can do this if you wish...but make sure the sound is impressive enough to have bragging rights to occupy that pan field. Most times keys and orchestra type stuff will be filler material...backing instruments. It's nice to have a handle on them so that they don't run amuck in the mix. The reason I say this....a stereo synth that is hard left/hard right, will nearly sound centered when you really listen to it. There we go again adding more stuff to the phantom center pan field, understand?
 
Stereo imaged sounds (which is just what a sound is when you recorded it with dual tracks at once) is not the same as a simultaneous, individual take. THAT is where true stereo comes in...and then you lose the "phantom center". But with stereo takes which include keys, guitars from amp sims, processors etc....anything that is not played 2 times, will always be stereo imaged. For example, lots of guys are under the impression that if they plug into their POD's on guitar and run two outs stereo, they are recording in stereo. This is not true. The only thing in stereo will be the effects if any are used. Chorus, delay, flange, verb, etc. Distortion/over-drive is not a stereo effect. So anyone recording those dual stereo tracks at once that look like this:
 
▄▄▄▄▄
▄▄▄▄▄
 
Is not recording in true stereo. It's not a bad thing, but it's not true stereo and you'll get much cooler stuff by double tracking in my opinion. Even with keys...the waves within the sound are moving left to right...it's stereo imaged, not true stereo. If you want it to really rip your face off, record it twice and let your human inconsistencies put it in true stereo...and listen how cool it will sound when the waves within the synth move left and right along with your new performance. 
 
But for vocals, to me less is more unless the back up vocals are where you are going to be adding your layers. For example, I had mentioned to you the other day that I was working on a vocal part for the CHB. The hook in the chorus line in the song is a back up vocal that simulates a country singing. I sang the back up harms 25 times and added Clone Ensemble to it...it's crazy sounding, but it hits VERY hard. For a vocal like that, we have 2 options. We let it consume the entire mix left to right like an entire country of people singing would do, or we tighten it up and allow it to be slightly wider than the hats and ride cymbal. Both methods work VERY well and the message still gets through. As a producer I have to decide if it's worth it to "tip my hat" and lash out with this in those hard L/R pans or, do I keep it tight? What is better for the song? In this case, I don't know yet because the song isn't done and this will depend on what the other instrumentation is doing.
 
So hopefully some of this helps you or at least gets the wheels spinning for you. Multiple vocal tracks are cool to experiment with. If they are lead vocal tracks and you have several, the best bet is to try and come up with a way to make them an entity so they don’t sound like a bunch of lead vocals fighting for the throne, ya know? I’d have to hear what you have going on to really comment further. So when you’re ready, let me know and I’ll give it a listen for you. Uggh…another novel. Sorry Philip…if only I knew how to be less wordy yet keep the information factor. I just hope some of it helps you man. Best of luck.