I've been experimenting with different techniques for vocal doubles (hip hop): squash em, parallel compress, smiley EQ, boost mids, etc.

I can see benefits to all approaches depending on the scenario, but I'd like some input from others. For instance, the smiley curve EQ is nice for a more mellow track where you want to add some dimension to the vocals, but not necessarily forward enough for a more energetic track. I like parallel compression (or at least more modest if 100% wet) in almost all scenarios. Heavy compression seems detrimental to the vocal playing well as a background piece.

But look - I'm still pretty new to all this. My ears are getting better every day, but I really want to know what you guys do and more importantly what your choices are based on.

 

@ Cav Emp,

Unfortunately, I don't be myself purely involved in 'Hip Hop Vocals', so I had to search the exact definition of 'Vocal Doubles'.
According to 'Wikipedia' : "Double tracking or vocal doubling is an audio recording technique in which a performer sings or plays along with their own prerecorded performance, usually to produce a stronger or "bigger" sound than can be obtained with a single voice or instrument. It is a form of overdubbing."

I have a suggestion, even if it's not so 'glorious' as can be the fact to directly 'doubling' your first recorded Vocals with your own Voice, but you can get interesting results with it. *yes*

After recording your first 'Take' ('Track') with the desired Vocals, simply 'duplicate' that recorded 'Take' ('Track') to another empty one Track (Channel), then simply move (shift) that duplicated Audio Material from a few 'ms' ('miliseconds') compared to the original one, then 'pan' each one according your to needs, wishes,... *yes*
You'll get the effect of 2 Singers ('doubling') singing at the same time, and as both 'Takes' ('Tracks') are slightly shifted, you'll get a really 'natural' and 'human' touch... *yes*

About :

For this, maybe you should take in account the 'Frequency Spectrum' of the Vocals you recorded (e.g. 'Male', 'Female',... ) and act ('EQing') accordingly... *yes*

This kind of Compression ('Heavy') is perhaps more indicated to use on some Instruments (e.g. 'Drums', 'Kicks', 'Snares',... ), than purely on 'Vocals', apart if you're looking for some 'special' effects...

And finally :

You're lucky ! Because as you become older is rather the opposite that happens !!!


And this individual was far from being an illustrious unknown...

 

Not necessarily pertinent to hip-hip, but here are some things I consider with doubled voices, sung or spoken (or both):

• Tight harmonies panned to the same place (i.e., center); contrapuntal voices/lines panned separately.
• On both-centered voices, slightly different EQ settings on either, for differences their in characteristics, especially for harmonies in different registers (for example, more low-mid (or less high-mid) for lower; the opposite for higher.
• Gentle compression, leveling or limiting on either voice, sidechained into common compression for both.
• One voice recorded close to the mic; the other recorded from at least arm's length away.
• One voice singing (or speaking) softly; the other more loudly (though not much).

I suppose that I'm envisioning a "funnel" in which two elements come out in equal but distinct proportion - like a vanilla/chocolate swirly frozen-custard cone. (I think I'll go have me one right now, in fact.)

 

Through experimentation you will find that all the above suggestions might work in the context of your mix. Another tack you might experiment with is using a delay, reverb or chorus to create a slightly different space for one of the voices when they are both center panned. Everything and anything could work for you. Have fun as you try out and I am sure that you will be awed at your outcome. And of course please share when you are done, curious minds and such.

 

There's this thing in Eventide Ultrachannel called "micro pitch shift", can someone more knowledgeable explain how this can be achieved without it or what's the idea behind it, sorry for hijacking this thread, it seems to me this is some neat stuff for vocals

 

Oops. Guess I forgot this forum is primarily EDM/electronic music and almost completely European.

I've tried shifting an actual double of the track. It gives the desired effect and sounds very clean. However, sometimes I feel it's too perfect and people who listen to rap will recognize it right away, but the beats I make are heavily influenced by certain EDM styles, so I guess I can ditch a genre convention or two (lol) and try it out, see if I like it.

There is always stuff like Waves' Reel ADT which does a similar thing and even shifts the copied vocal with a random LFO so that it doesn't sound so cut & paste. It's nice for when you want a subtle double through the whole verse/hook/whatever, but it only gives you one copy so you can't really keep the main vocal center and pan one left, one right (maybe with a sidechain).


@Stevitch - recording farther away is a great idea! That's one I hadn't thought of yet. I've done every variation of recording a double softer/louder/hyper/blah. I've gotten some REALLY cool results with that trick. Especially a sort of hype track with a deadpan vocal double underneath.

@superliquid Agreed, although I'm kind of weak with reverbs. It's one area of mixing/producing I kind of ignored until recently

 

Lucky for you, Im not possessive about my threads

But I don't know anything about Eventide Ultrachannel

 

yeah dischord from audio damage preset subtle doubler with some tweaking

OT:
On more then 1 occasion I mute the doubles, in some strange way rappers always want to double their raps to get that FAT sound. I always comply to their request and use the best
(this is an international problem since I work with artist over the border) not everything fits in different situation is my opinion.

but to ad add one taste,

1. hard pan 2 (ad-lip/doubled) low volume lines, put them in a group to make it stereo.

2. split it in 2 tracks : 1 clean track and 1 track with ping-pong delay with a short feedback and delay time with the dry signal muted

3. then give the delay track some drive/distortion and or cut the highs and lows, and turn the volume all the way down till it blends in

then put the 2 tracks together back on the master group of that rap vocalist and ad a little extra compression

I can always prepare some interesting mixer presets in fl studio..
grtz

 

You are better man than I am

 

It's not just about fat sound, IMO. If the track has a hard vibe or a lot of energy, one dry vocal just doesn't sound right sometimes - especially if the rapper isn't, say, DMX. On hard/hype tracks I usually have one barely audible double throughout the verse, either separately recorded or with something like a chorus/ADT sitting mostly (or fully) center with the main vocal. Then I'll bring in two more layers on parts that want emphasis, each panned 35-70% left and right. Hard-panning works for the hook or when you need something to be WIDE, but when I just want to beef up a vocal I think it sounds better to leave some in the center.

I hadn't thought of a ping pong delay for achieving the panned-doubles sound. Def gonna try that one out.

I normally do add a bit of overdrive/tape sat on doubles, but I hadn't tried a high and low cut. What I've gathered is, when you want something kind of whispy to just add a little bit of dimension, cutting the mids seems to work pretty well. When I want some real forward, beefy doubles, boosting around 1.5k works nicely. Just gotta make sure you (or in your case, your artist) is precise enough to match his own rhythm. If you boost 1.5k on a separately recorded double where the rapper is even slightly out of sync with the original vocal, it's gonna sound super sloppy. Man I hate sloppy doubles, and there are mainstream tracks where these dudes can't reproduce their own flow. Shit's pathetic.

Thanks for giving me some things to think about and try out, man!