question about stacking vocals

[fanito]

hey dobro, my guess would be abascus just meant the second time around was focused on articulating consonants and words that didn't sound all that clear but then i'm just guessing. someone suggested a whisper track which sounds kind of like a similar idea and i tried that on a part that was really hard to duplicate cuz it was quick - almost like a rap- and it worked well to make the words clearer. ciao, fanito... p.s. i'm still wondering about bouncing tracks. is it hard to line them up? and what is the "slight delay" ya'll are talking about? sorry i'm still a newbie...

 

[abuscus]

"I recorded this one singer and he had the idea to do a second vocal track where he just sang pronunciations to make the words stand out more"

Abuscus - what do you mean by 'pronunciations'? There are pronunciations every time you open your mouth and sing or say something - that's what singing and speaking are - pronunciations. What did the singer do differently on that second take?

Sorry,instead of saying he "sang pronunciations" I should have said he "whispered syllables".One guy called it a whisper track.It is a method used to make words stand out.Next time I'll be more careful what I say.It makes us feel smart to correct other people does'nt it?I am aware of what the word pronunciation means,thank you.Everybody else got the jist of what I was saying.Fucking english teacher over here!What's your insecurity?Talk to you later Webster!
P.S. Fanito,Im glad you had success stacking vocals.

 

[sjoko2]

Next time I'll be more careful what I say.It makes us feel smart to correct other people does'nt it?I am aware of what the word pronunciation means,thank you.Everybody else got the jist of what I was saying.Fucking english teacher over here!What's your insecurity?Talk to you later Webster!

Ever considered incorporating the expression "he's got a chip on his shoulder" in your teachings?
If no......... look inside self, and on own shoulder, for all required study material

 

[Titanship]

so if you recorded to two tracks...one with a slight delay, would you then bounce them together or leave seperate?

If you have enough tracks to spare, I would leave them seperate. That way you can either pan them both middle- or pan differently. That might give a neat effect. Of course, if it's the lead vocal, you probably won't want one in the middle, and one to the left or right. Also, if on seperate tracks, you could eq them slightly different for another "effect."

titan
peace

 

[Michael Jones]

You guys are making this harder than it has to be.

Record your vocals in mono on one track.
Copy that track.
Delay the copied track on the order of miliseconds.
Pan the original ever so rightish.
Pan the copied track ever so leftish.

Mix well.
Season to taste.

Good recipe for great vocals.

 

[littledog]

.Next time I'll be more careful what I say.It makes us feel smart to correct other people does'nt it?I am aware of what the word pronunciation means,thank you.Everybody else got the jist of what I was saying.Fucking english teacher over here!What's your insecurity?Talk to you later Webster!

You may want to remember, abuscus, that there are many non-native english speakers on these boards who may not have the same fine intuitive grasp of the language that you have. Then there are morons like me who also have trouble guessing what it is you are talking about when you butcher the language. Speaking for both groups, we are both deeply apologetic that we weren't able to instinctively understand that by "pronounce" you meant "whisper", and that one of us had the audacity to ask you to explain your meaning.

 

[littledog]

You guys are making this harder than it has to be.

Record your vocals in mono on one track.
Copy that track.
Delay the copied track on the order of miliseconds.
Pan the original ever so rightish.
Pan the copied track ever so leftish.

Mix well.
Season to taste.

Good recipe for great vocals.

With all due respect, Michael, the simplest solution really is:

Sing the song a second time on a second track.


All the copying, delaying, panning, Eventiding, chorusing, detuning, etc. in the world still won't sound as good.

 

[Michael Jones]

With all due respect, Michael, the simplest solution really is:

Sing the song a second time on a second track.


All the copying, delaying, panning, Eventiding, chorusing, detuning, etc. in the world still won't sound as good.

Well copying a track on a DAW takes about 2 seconds.
Delaying the start point on the copied track - about the same.
30 seconds for even an idiot to pan.
Eventiding, chorsing, and detuning, etc. sounds like over processing to me.

Littledog - I wont refute your technique as plausable, but in my findings its nearly impossible to get the vocalist to sing an identical track, i.e. in time, in tune, and on cue, to match with the first one.

Given that, you're left with 2 tracks that don't quite match and when played together sound like... well, like 2 tracks that don't quite match.
Maybe I'm missing something here?
If you'd care to enlighten me, I'll listen.

 

[Sky Blue Lou]

Hi Mike, pardon the interruption...

Hey littledog, how are you? I'm standing in line to get the answer to Mike's question (actually I guess I'm butting in ). I don't sing much, or well, myself but I use Mike's method to double (triple, quadruple...) guitar bits. It's not likely I could play the notes the "exact" same way twice. Is accuracy not a critical factor?

lou

 

[littledog]

Mike, I think you're missing the point. The fact that the performances are not going to be 100% identical is exactly what makes it sound like a real chorus. Let's say you have a group of four backup singers singing a part in unison. If they are good singers, they will phrase together, and the performance will sound tight. Maybe a conductor will even be utilized to help. But they will not be 100% in unison on their vibratos, cut-offs, accents, and other articulations. This is what makes it sound real. And what makes it sound good.

You can argue that it is too difficult to resing a part, but you are flying in the face of history - from the Beatles to your favorite pop artist of today. When you hear four Jennifer Lopez's singing the chorus of a song, that IS her singing the song four times.

Admittedly, it helps if you are a pro singer, because you want the phrasing to be as close as possible. But it will never be 100%, and that's a good thing.

Now, there ARE performances, like jazz improvisations, that really can't be duplicated. In those cases usually double tracking is probably not the sound you want anyway. But if you want to chorus a burning jazz solo, that would be the time to try some electronic tricks. BUT IT STILL WON'T SOUND AS GOOD as if you really played it twice. Check out the "Supersax plays Bird" for the sound of four-part saxophone renditions of Charlie Parker solos played in real time to hear the difference.

My point was not that you can record a second pass in less time than it takes to copy and paste. My point was that the copy and paste will never sound as good as a second pass, even if you have an Eventide Harmonizer etc. to help make the pasted vocal "different" with detuning etc. Since even if you spend hours with high end processors and fx it will never sound as good as having the vocalist just sing the damn thing a second time, then you are hardly "overprocessing". If anything, you aren't processing enough since your results remain inferior.

By the way, I won't be around much this week-end, so maybe we should wait until Dragon's new screaming server is up and running to continue this discussion...

Meanwhile, HAPPY THANKSGIVING all!

 

[Michael Jones]

Thanks for the response LD.
Maybe I'll try doing that again sometime. My results of trying that to date have yielded tracks too far out of time to be usable. And it doesn't have to be much out of time to render them useless.

I've had parts where it meshes nicely, but the consistency in time, across the entire take, wasn't close enough to be usable.

Anyway, thanks again.
Happy Thanksgiving for you and yours.

 

[object_play]

Peter Gabriel doubled his vocals often, and you can hear the fullness. I do not believe that Phil Collins did this ever since he took over singing for Genesis, and his vocal sound is clearly weaker. It makes you wonder if Peter Gabriels voice is really a better voice. He of course did much more grand stuff.

A pitch changer can be used also, just run the track through any multi-effects unit, and put it on one of the pitch phasing presets. It makes it a tiny bit fuller, but it's a little electronic sounding, and hardly compares to actually doubling vocals on 2 tracks, singing the lines twice.

I just joined this forum. Hi to all. I just bought a Tascam M2600 24 channel board to put in my home recording studio. It's the first good board I've ever had, and I'm excited to get it going. I also bought a Roland DM-800 8 track digital. I found it for $199, can ya imagine? Thing originally listed for 6 grand. I'd been using a fostex 8 track 1/4 inch analogue all these years, and synced up to SEMPTE track software, which I like. I had to use the 8th track for that, so only had 7. The machine is shot, as are my old junky boards. The DM-800 actually does it's own SEMPTE, so this should all be pretty cool. A rebirth. I'll probably be looking for advice on the M2600 and the DM-800 here pretty often. Stay cool, all.

 

[Garry Sharp]

Hi 2 U Object - new here myself.

Our singer is very accurate - when we went in a studio once the engineer pulled up two waveforms from consecutive takes, lined them up against each other and said he'd never seen such precise timing.

Anyway, last night I did the two take thing with her and it sounded horrible - like two girls singing in unison - fecking Bananarama or the Spice Girls. I still feel ill just thinking about it. When I've worked out how to do the few ms delay of a copied track on my brand new VF80 (fantastic box, unbelievably crap manual) I'll try that instead.

I have tried in the past adding chorus and suchlike using Logic on the PC and it just sounds so electronic.

 

[sjoko2]

You guys are making this harder than it has to be.

Record your vocals in mono on one track.
Copy that track.
Delay the copied track on the order of miliseconds.
Pan the original ever so rightish.
Pan the copied track ever so leftish.

Mix well.
Season to taste.

Good recipe for great vocals.

Good recepy for phase cancellation - check your mixes in mono and you might find your vocal has disappeared.

Personally, I agree with littledog. The only way to double a vocal is to sing it.
If you cannot get it 100% right, guess what. It takes work and practise to get it right. There are VERY few people (I have met 3 in 30 years) who are so good, and have such a fantastic memory, that they can match phrase-for-phrase perfectly.
So, if you have a studio at home, time shold be no constraint, work on it untill you get it done, which is the same a pro would do.

If you cannot do the above, or do not want to do it for sonic reasons, there are many processors available which you can use depending on your tast, from choral effects to reverbs, delayed reverbs etc. etc.

Oh yeah, Littledog? I downloaded the vocalign trial - and bought it. Good tool!

 

[Philby_Walsh]

check out...

Check Out Elliot Smith...he double tracks his vocals a lot..
can sound awesome sometimes but can also sound fake, like you're trying to cover up a poor voice..

abuscus - by 'slight delay' he meant add a slight amount of delay to the track.

 

[scottr]

Hi 2 U Object - new here myself.


Anyway, last night I did the two take thing with her and it sounded horrible - like two girls singing in unison - fecking Bananarama or the Spice Girls. I still feel ill just thinking about it. When I've worked out how to do the few ms delay of a copied track on my brand new VF80 (fantastic box, unbelievably crap manual) I'll try that instead.

I would recommend bringing one of the tracks (the weaker one, whichever that means to you) down a bit to avoid a "double" sound.. This seems to work pretty good for me. I just punch in any parts that still sound off.


Scott

 

[Bigus Dickus]

Yeah...

So can I double track a classical piano performance to "fatten up" my piano sound?

 

[sjjohnston]

So can I double track a classical piano performance to "fatten up" my piano sound?

Whatever you do, don't double violin parts, particularly in classical music. God forbid you should actually get more than one different person to sit next to each other and play the same thing on several different instruments! That would sound crappy.

 

[Robert D]

Scottr has it right. Getting the two takes to be almost perfectly in time, and then using your slice and dice editor to fix the last little bits of the weaker take is the first and second step. Then the Dub Vox needs to sit under the Lead Vox, down to the point where it just becomes indistinguishable. You'll almost think you've gone too far and it's not there anymore, but hit the mute button on the Dub track and you'll hear it go away, and the Lead Vox will get thinner and one dimensional. Unmute and hear it thicken up and fill in the voice. It's not called for in every song, but done right it's pretty transparrent, and way better than simply using a delay, either by device or by track copy and skew.
To get the singer to get it close enough, try bringing up the 1st take pretty high in the cue mix, and don't send the Dub Vox track back into the cue mix, or send it, but mixed way down in comparison. Have the singer lift one side of the headphones off a bit to hear his voice while listening to the boosted previous take plus the mix in the other ear. Take it one verse at a time, or even one phrase at a time if you have to, repeating until it's right before moving on to the next phrase or verse. Get it to the point where it will be faster to edit the last little differences than to keep trying more takes.
It will never sound right if the two takes are the same volume, that's the key.
Cheers, RD

 

[littledog]

I don't think it's possible to get a more clear or eloquent explaination than that one.

Good work, Robert!