Vocal Recording: 1 take vs. punch-ins?

  • Thread starter Thread starter Jack Russell
  • Start date Start date

What method do you use most often for vocals?

  • The punch-in method.

    Votes: 142 58.0%

  • The entire vocal track is recorded in one take.

    Votes: 62 25.3%

  • If the singer fucks up, I zap him/her/me with a cattle prod.

    Votes: 41 16.7%
  • Total voters
    245
I sing it over and over and over and over until it's right in one take. Or put if off until another day. Mainly because I suck at punch-ins.
 
It really helps me to record a guitar playing the melody I'm going to sing, for a reference track. Then when I record, I sing with the reference track (which is later muted). You guys probably already know that trick, but I just finally discovered it myself, so I wanted to post it for the benefit of any other newbies here. :)
 
I always do four or five takes, pick the best performance and use the other takes for comping. I use the audio part editor in Cubase.

Cheers!
 
Hmmm...yeah....punch in's if I must, but not often at all. My best vocal tracks are ones that are multiple takes that are as identical as possible all mixed together into a single lead vocal track. Usually the "best" individual take would be on "top" with each individual other take being mixed into it at lower volumes and panned *a bit* L and R with light effects. Works for me.... :D :cool:
 
SouthSIDE Glen said:
And you know what? Sometimes a performance is right even if there are a couple "mistakes" in it. Perfection is not always perfect.

The only reason Loius Armstrong first started scatting was because he forgot the words. Imagine if that never made it to tape (or schellac).

And there's no law that says every last binary digit or oxide crsytal has to be perfectly clean of sibilance or popping. If it adds to the feel of the performance, leave it in. If not , take it out.

G.
Click to expand...

I think That is completely spot on, in this occasion. Only yesterday i was having a "disscussion" about so called "perfect" tracks. I believe that when i listen to my recordings where i know there are mistakes im focussing on those and not hearing the bigger picture. However when anyone else listens to the same track usually the listen to the whole thing rather than the little indiscrepancies in sound. Its these Little inperfections that make what you do unique, The difference between man and machine. You will notice it with drummers alot, especially if you check the crack of a snare, rarely to never is it identical the whole way through a song. A drum machine on the other hand is. Guess which one is more pleasing to the ear. I think what im getting at is that Performance is the key to a great record, if it is all punched in and out and at the same levels, over compressed and processed it can get boring fast. On the other hand theres alot to be said for punching in and out. What ever works for you is always right because only you can evaluate it and say whether it is close to what you want. It makes me think of an interview with George Martin where he says that he can hear the Join in "Strawberry Fields Forever" where they cut the Two versions and speeded up the tape. I imagine it kinda grates on him, and yet we are talking about one of the most innovative and important songs in terms of advancing music production.
If you go to a Puppet show and you are watching the strings, you really have wasted your money.
 
I would say do multiple full takes and cut up the takes from there taking the good parts from each and making one good take thats how a lot of the pros do it.
 
What I do with my singer is work section by section. Lets say we are working on the verse:

I will have him sing the whole verse (or if the verse is cut in half due to volume and tone differences then he'll sing the first half). Then if there is only like one noticable error we will correct it by punching it.

The main punching happens when we are working with layers. We will punch all sorts of parts in until the layers completely match with out wavering in pitch with each other.

This whole process takes much less time in my opinion and ensures that it will sound right when coming to mix it all together.

Other times this is not possible when a part calls for more of an emotional tone that isn't easy to replicate time and time again. That is more of a one take process.
 
I do each phrase/stanza as a punch in most cases. That allows me to concentrate totally on the 5 second segment I'm singing, because I need the help. ;)

Doing this, you pay a price dynamically, and you have to be aware to make all the takes soundthe same to keep it from sounding like patchwork. I end up heavily compressing my vocal anyway, so it sounds pretty even.
 
I used to have a terrible voice, I took a voice class and got to the point of passable, but still not very good. I just sang a song at a wedding and to get it right I practiced it at least once a day for 2 months straight, no joke. I went to record it yesterday and for the first time in my life I was amazed by my voice.

a few things that helped along the way...

I think one of the biggest problems with un-coached singers is that they learn by singing along and thus emulating different people's voices which doesn't help develop THEIR voice. STOP THAT! concentrate on your voice and do whats right and comfortable, eventually that will help a lot.

then the most important part of singing, hands down. breathing. take full deep breaths. of course you won't sound good without enough air. its like driving with an empty tank. finally sing with confidence.

disclaimer: I am in no way a very good singer, but I got from complete suck to pretty decent with some of these tips and practice.
 
yeah - when the singer allows and the situation calls for it...i do three takes at the same time...and comp after...

overdubbing separate days really never has worked for me too well...

i just listened to a book on cd on a long drive where the editing was HORRIBLE! like you could tell the takes were from completely different sessions. drove me nuts. :eek:

MIke
 
Most of the time, for my songs, I do 2 full takes, one after another. Most of the time, the 1st track is satisfactory, and the 2nd track is for doubling or pingback reverb (occaisionally).
 
Jack Russell said:
True. It is almost bizarre, isn't it?

I've read somewhere that the ears function like muscles. If they work too long, with loud sound levels, then the muscles get tired, and your hearing isn't 'true' anymore.

Maybe that is it?

Unless, of course, you were using gin for inspiration the night before. Then you can blame the booze. :eek:
Click to expand...

It's like walking into a building that stinks. If you stay long enough you begin to get used to it... and before you know it you don't even smell it anymore. Then someone else walks in and yells 'holy crap! what died in here?', and you're all 'what? I don't smell nothin'.'
 
I do one full take at a time.

I consider it a matter of pride, to be able to actually sing a song the whole way through, and sing it well and mistake-free.

But, hey, I love to sing.

I'll punch in back-up vocals if need be, though. But I really like being able to get through the whole lead line.
 
First off, I sit down with a lyric sheet and create a rough comp chart that follows the song structure. Then I'll create like 5-10 tracks in Pro Tools, and start from there.

I try and get as many takes as possible, making notes along the way of how each phrase went. After we've make some takes, I'll go back and listen to each take over again to decide which ones to keep. After that, I cut them up, and compile a new track with all of the edits.

My main issue is that most singers concentrate on pitch over timing. I'll take someone who can sing in time who has some minor pitch issues over someone who has no time.
 
I just set to loop record and sing though the whole song 3 or 4 times.
Find the best take and steal from the others as/if needed.

It's hard to match energy levels puching to me. And sometimes the first sing threw is just warm up.


F.S.
 
for normal rock vocals, i tend to punch in. Most singers aren't amazing enough to do it in one take. it's a lot easier for them to make the preformance solid that way, rather than trying to fix it with things like auto-tune later.

that way, when it comes to mixing, your job is closer to being done.
 
I always ask my singer to record 3 to 4 times, then I will mix after that, its easier to mix if there's any mistake in the singing.
 
Does nobody record several takes and blend the best passages from each take?
 
My vocalist is strong enough to usually get it in one take, but usually i do 2 or 3 takes and comp it.
 
You must log in or register to reply here.
Back
Top