Mixing Doubled Vocals

Recording Techniques, People Skills, Gear, Recording Spaces, Computers, and DIY

Moderators: drumsound, tomb

User avatar
losthighway
deaf.
Posts: 1773
Joined: Mon Apr 14, 2008 8:02 pm
Contact:

Mixing Doubled Vocals

Post by losthighway » Thu Dec 06, 2018 8:38 pm

"Hey, let's double that lead vocal!" Says so many people who come in to my studio.

The thing is, as a listener I generally don't enjoy hearing this. Sure, Elliott Smith or Cat Power, maybe Kurt Cobain got it to do something additive, but I generally don't like it. But I always oblige anyone who feels like they dig it.

Usually I compress the one we like best slightly more and have it up a little louder, roll off a little high and keep the secondary one a little lower. This tends to put it closer to a slap echo feel depending on how well they match.

*BUT*

There are so many ways of presenting this common thing, and I thought it'd be smart to learn about a few more from some of you.

User avatar
A.David.MacKinnon
ghost haunting audio students
Posts: 3115
Joined: Wed May 07, 2003 5:57 am
Location: Toronto
Contact:

Re: Mixing Doubled Vocals

Post by A.David.MacKinnon » Thu Dec 06, 2018 8:57 pm

Equal volume, hard panned left and right?
Crush the shit out of on and sink it low under the other? And/or use it as the feed for delay and reverb?
Mangle it and with whatever and then feed the mangle into an H3000 doing the slightly pitch shifted, slightly delayed L-R thing? Return it 100% wet with no dry, mangled signal.

drumsound
zen recordist
Posts: 6563
Joined: Tue Jun 01, 2004 10:30 pm
Location: Bloomington IL
Contact:

Re: Mixing Doubled Vocals

Post by drumsound » Thu Dec 06, 2018 9:06 pm

Generally, I pull up the fader on the double kind of slowly to find a good blend. Sometimes it ends up about equal, sometimes way behind. Usually I''ll only send one of them to reverb or delay (if I'me using them on the vocal). Occasionally the double makes for a little bit more interesting send than the main vocal, usually not.

kslight
resurrected
Posts: 2417
Joined: Tue Oct 13, 2009 7:40 pm

Re: Mixing Doubled Vocals

Post by kslight » Fri Dec 07, 2018 7:21 am

I don’t love straight unison doubled vocals. Generally I’ll pick a lead track and do normal lead vocal processing with that, then guide the double to that (editing wise), often attenuating or removing breaths and t’s and s’s completely, and shortening the tails of phrases so you don’t get that annoying moment where they don’t line up at the end of every line. I also may band pass that double, not a lot of low or high end, and maybe throw a teeny amount of drive on it. Blend to taste (just enough to notice it when it’s gone, usually). Might also add some other kind of processing to the double, like chorus or shimmer or just anything to differentiate from the lead...keeping the lead track relatively clean.

Magnetic Services
pushin' record
Posts: 268
Joined: Mon Apr 16, 2012 11:21 pm
Location: Milwaukee

Re: Mixing Doubled Vocals

Post by Magnetic Services » Fri Dec 07, 2018 1:19 pm

I did a record recently where the singer doubled or tripled (or more) almost all the vocals on every song. On some songs I was able to convince them to start with a solo vocal in the verses and add the rest in the choruses to give it more power, but on some songs I had to be sneaky, so I came up with a technique:

I would stack the vocal tracks in mono and emphasize the best one, so it sounded thick but didn't lose too much focus, then in the choruses I would automate the panning to make it nice and wide, either L/R or L/C/R. It's a win/win: the singer gets to hear their voice triple-tracked, and you get to "arrange" the vocals the way you want with volume and panning.
Dante Fumo
Engineer and freelance audio writer
dantefumo.com

User avatar
vvv
zen recordist
Posts: 8439
Joined: Tue May 13, 2003 8:08 am
Location: Chi
Contact:

Re: Mixing Doubled Vocals

Post by vvv » Sat Dec 08, 2018 11:12 am

For me, recording myself, I think of it as an arrangement tool. My typical approaches are:

a. a straight double that is near-exact and used to create that straight doubled sound, which in the mix might result as being used all the time, or just on the ends of phrases, or the ends of the line, or only on some lines or verses.

b. sing it in a different (generally lower) register, and then mix as above.

c. rarely do I try the "Heroes" thing where it's one performance, recorded two different ways, and then used in parallel and/or as an alternate.

That said, I almost always double the chorus as a. or b. or a. & b. above, because that's my "sound".

Note: then there's the John Lennon, & Ozzy, ADT thing ...
bandcamp; vlayman;
THD; Geronimo Cowboys;
blog.
I mix with olive juice.

Magnetic Services
pushin' record
Posts: 268
Joined: Mon Apr 16, 2012 11:21 pm
Location: Milwaukee

Re: Mixing Doubled Vocals

Post by Magnetic Services » Sat Dec 08, 2018 1:03 pm

vvv wrote:
Sat Dec 08, 2018 11:12 am
then there's the John Lennon, & Ozzy, ADT thing ...
do tell...
Dante Fumo
Engineer and freelance audio writer
dantefumo.com

User avatar
losthighway
deaf.
Posts: 1773
Joined: Mon Apr 14, 2008 8:02 pm
Contact:

Re: Mixing Doubled Vocals

Post by losthighway » Sat Dec 08, 2018 1:05 pm

Magnetic Services wrote:
Sat Dec 08, 2018 1:03 pm
vvv wrote:
Sat Dec 08, 2018 11:12 am
then there's the John Lennon, & Ozzy, ADT thing ...
do tell...
I think it's really just a super quick delay with a little bit of flange/pitch-shift.

User avatar
A.David.MacKinnon
ghost haunting audio students
Posts: 3115
Joined: Wed May 07, 2003 5:57 am
Location: Toronto
Contact:

Re: Mixing Doubled Vocals

Post by A.David.MacKinnon » Sat Dec 08, 2018 3:08 pm

losthighway wrote:
Sat Dec 08, 2018 1:05 pm
Magnetic Services wrote:
Sat Dec 08, 2018 1:03 pm
vvv wrote:
Sat Dec 08, 2018 11:12 am
then there's the John Lennon, & Ozzy, ADT thing ...
do tell...
I think it's really just a super quick delay with a little bit of flange/pitch-shift.
The Ozzy thing is a really quick delay de-tuned by 10-15 cents or so. Same goes for the Jane’s Addiction sound.

ADT is another beast entirely. Reading about it in the recording the Beatles book hurt my brain. It seems like if it’s implimented properly the delay time is always changing and can sometimes arrive before the clean vocal. Or I’ve got it wrong. Either way, being the best band in the world obviously involves bending the rules of space and time.

User avatar
vvv
zen recordist
Posts: 8439
Joined: Tue May 13, 2003 8:08 am
Location: Chi
Contact:

Re: Mixing Doubled Vocals

Post by vvv » Sat Dec 08, 2018 6:40 pm

See the wiki entry for some info.
bandcamp; vlayman;
THD; Geronimo Cowboys;
blog.
I mix with olive juice.

User avatar
vvv
zen recordist
Posts: 8439
Joined: Tue May 13, 2003 8:08 am
Location: Chi
Contact:

Re: Mixing Doubled Vocals

Post by vvv » Sat Dec 08, 2018 6:49 pm

Then there's the Alice in Chains thing where Cantrell could almost exactly mimic Staley's voice, and they sang in unison.

I amuses me on the last cuppla albums with the new singer guy, who is singing close to Staley's voice and Cantrell's mimicking him, except sometimes I think it might go the other way.
bandcamp; vlayman;
THD; Geronimo Cowboys;
blog.
I mix with olive juice.

User avatar
losthighway
deaf.
Posts: 1773
Joined: Mon Apr 14, 2008 8:02 pm
Contact:

Re: Mixing Doubled Vocals

Post by losthighway » Sat Dec 08, 2018 7:59 pm

A.David.MacKinnon wrote:
Sat Dec 08, 2018 3:08 pm
ADT is another beast entirely. Reading about it in the recording the Beatles book hurt my brain. It seems like if it’s implimented properly the delay time is always changing and can sometimes arrive before the clean vocal. Or I’ve got it wrong. Either way, being the best band in the world obviously involves bending the rules of space and time.
According to this guy it's delay plus phase/flange.

https://www.youtube.com/watch?v=TgnSVdjfSwk

User avatar
A.David.MacKinnon
ghost haunting audio students
Posts: 3115
Joined: Wed May 07, 2003 5:57 am
Location: Toronto
Contact:

Re: Mixing Doubled Vocals

Post by A.David.MacKinnon » Sat Dec 08, 2018 10:44 pm

Thank V! From the wiki -
“When mixing a song, its vocal track was routed from the recording head of the multitrack tape, which was before the playback head, and fed to the record head of the second tape recorder. An oscillator was used to vary the speed of the second machine, providing more or less delay depending on how fast or slow the second machine was run relative to the first. This signal was then routed from the playback head of the second machine to a separate fader on the mixer. This allowed the delayed vocal to be combined with the normal vocal, creating the double”

So, in theory if the delay machine is running faster than the multitrack it’s possible for the double to arrive before the original. It still hurts my brain.

User avatar
losthighway
deaf.
Posts: 1773
Joined: Mon Apr 14, 2008 8:02 pm
Contact:

Re: Mixing Doubled Vocals

Post by losthighway » Sun Dec 09, 2018 5:26 am

^ Yeah, there's some nitty gritty about the mechanisms of tape machines there that I'm not sure of. I don't think it's actually physically possible for the secondary signal to arrive any sooner than simultaneous to the original, unless record head sent to alternate record head still comes out ahead of original playback head. It's a game of very few ms, but what you're describing may be possible. Maybe one of our analog recordists will chime in here.

As to the varying delay with an oscillator, this is really the first example of a phaser/flanger. We can set up a mimic effect with an Effectron, or a multifx box, or a couple plugins. Maybe not quite the same as a couple of tape machines, but you can achieve a similar flavor.

User avatar
A.David.MacKinnon
ghost haunting audio students
Posts: 3115
Joined: Wed May 07, 2003 5:57 am
Location: Toronto
Contact:

Re: Mixing Doubled Vocals

Post by A.David.MacKinnon » Sun Dec 09, 2018 6:17 am

If the feed to the delay is coming from the record head of the multitrack but the mix is coming off of the playback head then there will be a delay between the send and mix signal. At 15 IPS the send will be about 80ms earlier than the playback/mix signal.
If the delay machine gets the signal 80ms early and is running faster then 15 IPS the the double can get to the mix channel before the original signal.

Post Reply

Who is online

Users browsing this forum: No registered users and 28 guests