Layer a sampled choir along with the multiple real voices. 

I have used clone ensemble to good effect but also got some pretty good results with Antares Harmony Evo using the choir mode (they also do Choir EVO as a separate effect).
 
I tend to mix it with real vocals for a better result. I will create a 4 part harmony (real) then layer various versions of Antares Harmony Evo (e.g. lower, mid and upper registers controlled via MIDI) around it to create a much bigger sound. You can record each layer using an Aux track so that there is not much of a CPU hit.
 

Just coincidentally ran across this video today and it made me think of this thread.
 
Yeh, it's just an ad for Waves, isn't a direct answer to the original post, nor does it reveal any arcane secrets that everybody here doesn't already know.
 
But it's still pretty cool to see this demonstrated by someone who's well known for that wall-of-sound effect. If you've ever listened to any of Devin Townsend's more epic material, you already know that he's a master of the technique.
 
The key lesson here is that even though he's hawking effects for Waves, he doesn't synthesize any of the vocal parts with doublers or the like. Instead, he takes the time to actually record gobs of individual vocal tracks (all him, which I found surprising) and mix them into to a grand mush.
 

That's a cool video. Interesting to see someone else showing their techniques. 
Some of that is what I use in the effects dept.. 
Things I've done - I used different mikes,, different distances, sing in as many harmony parts as I dare.    but mostly it was just me overdubbing to 12 / 16 tracks and using delay and reverb to make it sound like more than it was.. 
 
If you have other people then that is even better. I had 6 people come in and we had to feed them Tequila to loosen them up. We called it the Tequila Angel Choir. We made 2 takes and mixed those together... This was in my 16 track days.  
 
 

It's always good to get someone else's voice in there, even if they're not a good singer. You're trying to assemble a rich mix of overtones and formants, which is next to impossible when it's the same voice doing every part.
 
When my granddaughters were 11 and 12, I recorded them singing some oohs. They were shy so it took a lot of prodding to convince them to do it. But after tuning and treating the tracks, they sounded like a choir of angels and were pleased with themselves. I had them each sing the same notes of the triad, six tracks in total, each treated and panned differently. It's similar to the technique used on Bohemian Rhapsody, which had three takes of three singers, who swapped each others' parts for each take.
 
I'd read of the same technique being used to create a violin section with one violinist. They set up four chairs, which he rotated between while overdubbing the parts. Why not just pan each part? Because when it's acoustical instruments, it's not just about panning; the room sound is different from one position to the next.
 
Which makes me wonder if the same trick might not work for vocals. Not necessarily moving around the room, which wouldn't help since you're probably recording in some kind of dead space. But you could vary the distance and angle from the microphone for each pass, or even change out the microphone, thus getting a unique tone for each take.

bitflipper
Just coincidentally ran across this video today and it made me think of this thread.
 
Yeh, it's just an ad for Waves, isn't a direct answer to the original post, nor does it reveal any arcane secrets that everybody here doesn't already know.
 
But it's still pretty cool to see this demonstrated by someone who's well known for that wall-of-sound effect. If you've ever listened to any of Devin Townsend's more epic material, you already know that he's a master of the technique.
 
The key lesson here is that even though he's hawking effects for Waves, he doesn't synthesize any of the vocal parts with doublers or the like. Instead, he takes the time to actually record gobs of individual vocal tracks (all him, which I found surprising) and mix them into to a grand mush.
 

Very interesting. Thanks for posting.

I use waves doubler and reverb. Sometimes I use melodyne but doubler does a great job on cloned tracks

Ironically, it is impossible to find this (true or false) but 30 years ago, a DJ (probably Casey Kasem) said that Hysteria's backing vocals were 40 tracks per member. Given that Mutt Lange is a brutal perfectionist in the studio (multitracks parts regularly) and the band admitted they spent up to two weeks tracking a single verse, I never disputed it.

As mentioned previously, a clone will only get you so far, so multitracking enough to build harmonies is prudent and will give more texture. The best harmonies are not static intervals, so most harmonizers fall short if they lock to only one interval (3rd, 5th, etc.), which is something to keep in mind, but ultimately it falls into the context of the piece and how present they are in the mix.

If the song has identical parts, esp. chorus, copy versions of all choruses over each other.
 
Sing some parts yourself, shaping your mouth for each part to get different formants. Embarassing, but you'll get used to it.

mettelus
Ironically, it is impossible to find this (true or false) but 30 years ago, a DJ (probably Casey Kasem) said that Hysteria's backing vocals were 40 tracks per member. Given that Mutt Lange is a brutal perfectionist in the studio (multitracks parts regularly) and the band admitted they spent up to two weeks tracking a single verse, I never disputed it.

I've done a lot of reading (and studying) about the making of "Hysteria" and everything you mentioned is true
(except the 40 tracks per member). Im not saying that isn't true, im just saying I cant verify that.(wouldn't surprise me though). They were aiming to make the next Michael Jacksons Thriller only in the rock genre.
 
What caught my interest was the use of the Tom Schulz Rockman's (I own multiple Rockman's). The whole album was made using exclusively Rockman's. No amps, everything was run direct. Bass and yes. Even Drums.
 
They were so perfect making this album that it should have been an over polished flop. and it almost was if they hadn't released "Pure some sugar on me". There is actually one part in the chorus of the song: Hysteria where they broke up all the notes of the chords on to separate tracks so Mutt could have total control over each note at Mixdown.