Get two or more solid takes and blend them.

Bodhi, post: 375044 wrote: Get two or more solid takes and blend them.

This. So much better than using a plug in but it can depend on the genre. Interesting hearing Brian May talk about how Freddie Mercury would double track like this and could create an almost phase problem-free second track if he wanted. However, often he would sing it slightly different on purpose to add more layers and "color" to the end result.

[[url=http://[/URL]="http://multitrackma…"]multitrack master | studio tracks, isolated[/]="http://multitrackma…"]multitrack master | studio tracks, isolated[/]
studio tracks from your favorite artists, isolated

The takes should be IDENTICAL or else INTENTIONALLY different. AKA harmony. Otherwise, it may sound campy or otherwise amateur. It can sound great if done right or awful is done poorly. If you are going to do it, be prepared to spend some time getting it right. Close won't be good enough. If it doesn't make you think "yeah that sounds awesome", it's not there yet. Take the time. Do it right.

Josh1115, post: 374044 wrote: The method that I normally record is basically have two or three tracks with segments of a vocal performance broken up and take the best take of that particular segment

This is essentially the way it's done. Pros can't afford to have mistakes so comping is the norm. With millions at stake, there can be no flaws.

The first main take of the vocals needs to be as near perfect as possible or you'll never get the second or third take even close. Many times while double tracking a vocal part it becomes apparent that the 1st take is not good enough. At that point you really have to start over, but even that can help you get to the point of near perfection. Keep repeating the process till it is done as good as perfect can be in your own terms.

In Cubase I'd make a harmony by duplicating the track and then transposing the duplicate using sth like VariAudio amongst other FX if necessary.

Crye, post: 401572 wrote: In Cubase I'd make a harmony by duplicating the track and then transposing the duplicate using sth like VariAudio amongst other FX if necessary.

Not the same result. Not nearly the same.

First of all, harmony doesn't always work like that, in fact it rarely works like that. You can't just knock one of the clone tracks down or up by "X" amount following the tonic vocal and expect it to work. Harmony vocal doesn't just follow the lead vocal in an equal increments note-wise.

The way to do it right is to sing the harmony part... be it something as simple as a block triad - or something more complex like suspensions, 7ths, 6ths, minor intervals, etc., with the original lead vocal take being as flawless as possible, because every other vocal part will be built around that original lead vocal track.

And finally....if you can't sing, or you can't sing harmony, please, for the love of God, get someone who can.

Absolutely agree with the importance of having an accurate reference track. One thing I've done - especially when working with untrained vocalists - is to take the lead vocal track (which might have been comped from a number of takes), copy to a scratch track, and process the hell out of it in Melodyne. Make the pitch dead on and the minimize any modulation or drift. It's not something that I would put on a record, but it's easy to sing harmony to. Record the backing tracks to the scratch track, then ditch the scratch track and mix the background vocals. Take the original lead vocal and see how it sits with the BG vocals. If it sits pretty well, you can use pitch correction in a more musical way to tighten things up. Of course, you can rerecord the lead vocals with the BG mix as well.

I should add that a well-rehearsed group of singers singing at the same time in the same room (screw the bleed and forget the Melodyne) usually sounds best - even if things aren't "perfect."

Bob...exactly! Just sad that putting that group together may not be possible at all times. And "you can use pitch correction" works for me.

Bodhi, post: 375044 wrote: Get two or more solid takes and blend them.

And eat the hard consonants on all but the primary vocal take.

old thread but close to my heart. this may sound dated and not suitable for more modern productions but here goes.

the way guys like Queen, CSN and Neil Young worked out was to do what were called vocal stacks. CSN would do these live with all three on a take and then triple that but essentially how its done by one person is to is to do one inversion of the harmony (say the 3rd) tripled (3 voices). then do the same for the 5ths any octaves (below or above) or variations. you can then take all these tracks and bounce them down to one or two tracks. this is the CSN "Suite Judy Blue Eyes" or Queen "Bohemian Rhapsody" sound.

with bands like portugal the man being pretty popular, and alot of hevliy reverbed out, lush vocals harmonies all over the airwaves, on alternative rock stations, the stacked approach is back in vouge. It's funny that with all these big artists doing records to tape, and live takes, pinpoint accuracy in almost imhumanly perfect vocal techniques like csn, or queen, aren't really given recognition as part of the 'classic' sound right now. seems like people only talk about the fidelity aspect of it.

actually i found with stacks that perfection isn't really required ... a little difference between takes can be a good thing ... rule of thumb is if it sounds good it is good ... as pointed out previously eliminating hard essses and consents on all but the main vocal is called for ...

kmetal, post: 405603 wrote: with bands like portugal the man being pretty popular, and alot of hevliy reverbed out, lush vocals harmonies all over the airwaves, on alternative rock stations, the stacked approach is back in vouge. It's funny that with all these big artists doing records to tape, and live takes, pinpoint accuracy in almost imhumanly perfect vocal techniques like csn, or queen, aren't really given recognition as part of the 'classic' sound right now. seems like people only talk about the fidelity aspect of it.

I love Queen but they were not always pitch perfect, but they were always musical. With autotuners, we too often chasing perfection and losing musicality. I have nothing against judicious use of studio tools like autotuners, quantizers and compressers but sometimes just the performance be. Listening to some of my favorite classic rock albums, I try to image how modern producers would have "perfected" the performances... To the point of sterilization!

Bodhi, post: 375044 wrote: Get two or more solid takes and blend them.

In order to make the takes even more tighter between them, google Vocalign Pro.
It really rocks and saves you hours of editing.