Recording a group of 20 singers....

L

Legolas1971

New member
Hi,

I'm going to be recording about 20 singers, take their vocals and
join it with an instrumental Mp3 file.

My question is: what would be the best way to do this? I was
initially thinking of recording them all at once in big room with
2 small condenser mics. However, this would give me less
control of each individual vocal track. Also, i would need 20
sets of headphones for them all to hear the Mp3 they'll be
singing too (that's just not doable).

Then I thought maybe I should record each singer individually in
my home studio. I built an iso-booth and recording each one
separately would give me more control of each singers'
performance....

Any other ideas?
 
Assuming this is a choral piece rather than a series of solos, what you gain in terms of controlling individual performances you'll certainly lose in terms of the way a good choir can weave their vocals together.

I'd "take the show on the road" and find a room with nice natural acoustics (a church or whatever). Then I'd mic the group with two SDCs in X-Y coincident pair mode. You'll need to play with the positioning of the singers and the mics--having two rows of ten and a riser for the back row might help.

As for playback of the backing track (why the heck is it MP3 rather than wave by the way?) a small speaker positioned behind the mics in their null area will keep the spill down to levels that shouldn't cause problems.

If there are any solos, I'd do those separately later. If your DAW supports convolution reverb, I'd do an impulse in the choir space in case you want to match the natural reverb later in the studio.
 
Bobbsy said:
Assuming this is a choral piece rather than a series of solos, what you gain in terms of controlling individual performances you'll certainly lose in terms of the way a good choir can weave their vocals together.

I'd "take the show on the road" and find a room with nice natural acoustics (a church or whatever). Then I'd mic the group with two SDCs in X-Y coincident pair mode. You'll need to play with the positioning of the singers and the mics--having two rows of ten and a riser for the back row might help.

As for playback of the backing track (why the heck is it MP3 rather than wave by the way?) a small speaker positioned behind the mics in their null area will keep the spill down to levels that shouldn't cause problems.

If there are any solos, I'd do those separately later. If your DAW supports convolution reverb, I'd do an impulse in the choir space in case you want to match the natural reverb later in the studio.
Click to expand...

Hey Bobbsy,

I should have explained further. This is a group of Vietnamese kids who
sing together (choir style) but they're not very good. There won't be any
"weaving" going on except for my ability to keep this thing together. So,
my idea was that recording them separately would allow me to fix pitch, etc...
There won't be any solos. Also, I'm gonna convert the backing track from Mp3
to wave and drop it into Pro tools....
 
I still suspect that a bunch of nervous kids will probably perform better in the security of a group rather than alone in a vocal booth in the unfamiliar territory of singing to music in headphones. By all means experiment but, when recording youth drama groups in the past, the biggest challenge in getting a result was keeping the nerves at bay. If they're used to singing as a choir my advice is to keep it familiar for them.
 
Bobbsy said:
I still suspect that a bunch of nervous kids will probably perform better in the security of a group rather than alone in a vocal booth in the unfamiliar territory of singing to music in headphones. By all means experiment but, when recording youth drama groups in the past, the biggest challenge in getting a result was keeping the nerves at bay. If they're used to singing as a choir my advice is to keep it familiar for them.
Click to expand...

Gotcha....good advice.....
 
Bobbsy said:
If your DAW supports convolution reverb, I'd do an impulse in the choir space in case you want to match the natural reverb later in the studio.
Click to expand...


Hey Bobbsy, what you said here really popped out to me. I understand what convolution reverbs are, and I use them, but how does one go about recording one in a real space? Is there a simple answer to that?

Thanks!
 
Well, the "simple" answer is "it depends on your DAW" but I suspect you want a bit more than that!

At the simplest, a impulse is just that: a recording of a brief sound and all its reflections/echoes in the space where you're working. However, there are a wide variety of techniques (and sounds) used to do this.

The two main ones are using an instantaeous noise or using a frequency sweep.

For the instantaneous noise route, something like a starting pistol is ideal but often frowned upon for some reason. The difficulties using a gun mean that quite often you use something like a film clapper board, a loud hand clap or even just popping a balloon.

For the second technique, a short duration frequency sweep (from, say 80Hz to, say, 15,000Hz) is played back via a speaker and recorded on a separate mic.

For either technique, there are various ways to do it but I'd typically generate the sound somwhere around the place where the performance is happening and record it on a mic something like half to two-thirds of the way across the room. I just use the same mic I'm using for the main recording, these days generally a stereo pair or MS stereo set up.

In my case, for ease, I tend to just use the clapper board/clap/balloon sort of thing but it can be argued that you can get a more accurate impulse reverb using the sweep. I've not had problems using the easy way and have been too lazy to do much playing!

A lot of this will depend on how your DAW wants to work and what sort of file they recommend--I use Audition so my response above is tailored to that.

Hope this helps!
 
Thanks! That's great help!

So then, you just keep a wave file of that small snippet and run it through the convolution verb right? Also though, how long should that file be? Just long enough to hear the decay of the room?
 
That's where different DAWs do things differently. In Audition, there's a command under the convolution reverb to "get impulse". It opens the wave file you select and converts it to an .imp file you can save in the reverb presets. Other DAWs may be different.

As for length, the length of you impulse to some extent controls the tail on you reverb. I tend to record until the echo has completely died but then (often) do a quicker fade out to shorten it.
 
You must log in or register to reply here.
Back
Top