you can duplicate the tracks and eq out the other singers, read about the latest black keys record

The mics that were used absolutely matters. Knowing the mics and their pick up patterns will greatly improve your chances of getting better results.

A cardioid mic pattern would provide a null (dead spot) behind each mic that you can certainly use to your advantage.

A condenser mic may be too sensitive to avoid picking up the other 3 vocalists, so a common dynamic mic like a Shure SM58 might be preferable to a fancy condenser under the circumstances.

Reflections of the other voices from a flat wall behind the mic might be a factor, so I'd consider putting the singers facing each other from 2:00, 8:00, 5:00 and 11:00 as an experiment.

And yes, they a variety of items to block mic bleed. We've had very decent results using the [[url=http://[/URL]="http://www.sweetwat…"]Auralex Xpanders[/]="http://www.sweetwat…"]Auralex Xpanders[/] with vocal groups such as yours in a large room facing one another. They just provide a little bigger dead spot behind the mic without getting boxy sounding. (They do affect your sight lines as well)

Best of luck!

Re:

Good way to control vocal breading.

For me recording an a'cappella group is like recording a quartet. It sounds different when you record them all together in the same room in a microphone then when you record them in separate microphones and mix separately. I would try 2 different methods.

1. Try a single mic in the room (Omni if available) and have everyone sing at the same time and voila. Simple and great sounding just like it is meant to be.

2. Use a pair of SDC's ORTF a few feet back and use some dynamic mic's like an sm-58, sm7, re-20 up close.

You'll likely get the most life like results with method one. You maybe able to enhance things easier with method 2 but far more likely you'll just get a mess.

I would over think this. Try option one. The group is the instrument not each voice.

Thanks for the quick response! I will definitely give these ideas a go-over and see what works best!

Paul999 raises a very good point. I would agree that a large part of the magic of a vocal group is often the natural interaction that happens when they all harmonize together. And if the room's sound characteristics don't get in the way, you can capture that with one good omnidirectional mic. The burden then falls on the vocalists to blend their parts through performance and/or mic position during the take to bring certain parts up and down in the final mix, because you won't have any other means to - for example, bring up just the tenor part in the chorus after it's recorded that way.

It's certainly a more organic approach, and sometimes the magic is in the subtle imperfections.

Try both ways and experiment, it's a learning opportunity if you can get everybody to play along. Maybe warm up your voices, in place with the recorder running.

But no matter which method you go with, make sure you're not sucking the life out of it in editing. If you're mixing and editing to overcome performance issues, in an attempt to simulate a perfect take, you might want to track it again paying more attention to the performance aspect of it. [complete with - Bob, you need to lean in a little closer for this part, and Larry you need to take it down a notch during that part]

Good luck. Keep us posted.

One of the most important things about working with bleed is making sure the bleed sounds good. If the bleed vocals sound tinny or boxy then it's a problem bringing them up in the mix. In you have three bad sounding bleeds, then that competes with the main source. If the bleeds all sound pretty good, then it's a question of level balance, not masking. So an alternative to Steve's suggestion above is to use eq to enhance the tonal quality of the bleed - rather than eliminate it. Seconds on the use of cardioid dynamics (e.g. SM58).

With all of that said, I think the suggestions about for using a stereo pair and/or an omni will give the best sound for a tight group.

dvdhawk, post: 381527 wrote: ....And yes, they a variety of items to block mic bleed. We've had very decent results using the [[url=http://[/URL]="http://www.sweetwat…"]Auralex Xpanders[/]="http://www.sweetwat…"]Auralex Xpanders[/] with vocal groups such as yours in a large room facing one another. They just provide a little bigger dead spot behind the mic without getting boxy sounding. (They do affect your sight lines as well)...

Any more details on these? I'm always suspicious when they sell a kit. Which pieces are good and which are not?

Bob, the one we use most often is the one that looks like a big bow-tie. The guy I work with on a lot of these projects has several of the complete kits. Based on the web description it must be the piece called the 14" Xpander. It's rather big and seriously can be a problem if they need to see over or around it for visual cues, but other than that - it's like putting a specialized little gobo where you need it most. All of my recording is still strictly mobile, so properly applied, we find they help minimize the bleed as well as the affect of these less than ideal spaces. We're usually using them in the exact scenario the OP describes - behind C414's, a TLM193 and similar caliber mics, recording vocal groups that need to sing together in fair to middlin' rooms. And if we weren't getting satisfactory results, we would absolutely be going the SM58 route. (Maybe I need to get a few more SM7bs so everybody feels like they're getting the studio experience)

Let me ask a question or two to see if I have an understanding of what you're referring to.

Are you hoping that you pick up each singer discretely enough that you can manipulate each channel independently (for things such as pitch correction, etc.)?
The individual effects you're trying to put on each track, are they individual compression, or are you talking pitch correction?

In general, recording a capella, you don't want nor need 4 completely isolated tracks. The vocal blend would be terrible on the final album. If you're looking for something more pop-ish like "Straight, No Chaser" or more classical like King's Singers, both should be approached in a very similar manner.

Here are a couple photos of a horn quartet whom I just signed to my label and released their Christmas album. These photos are from that session. The horn is a very "human voice" like instrument and I approached this in the exact manner that I would have for a vocal quartet.

Note, there's a central pickup mic and then 4 spots (the keen eye will pick up the 5th spot, but that's used for a quick vocal overlay).

I didn't add individual effects, nor did I pitch correct anything. The ensemble simply must be good enough that this is not necessary. Otherwise, you have to go to an individual tracking situation which sucks the life out of the project. (Though there are some pretty cool chord-based pitch correction tools out there, but I personally feel that working on ensemble blend and pitch are a way better investment).

For this group (and again, I'd do the same for a vocal ensemble), I used my main mic (a Royer SF12) as my main stereo feed. The spots were only used to bring up the clarity a little. The church is HUGE. I didn't want it to sound like the Grand Canyon!

Of course, if you'd like to hear the results of this recording...you'll have to visit my site and order a copy. ;-) It will be on iTunes in the next couple days.

Cheers-
Jeremy  

Attached files

Cucco - Are you hoping that you pick up each singer discretely enough that you can manipulate each channel independently (for things such as pitch correction, etc.)? Yes.
The individual effects you're trying to put on each track, are they individual compression, or are you talking pitch correction? - Just for pitch correction purposes.

Let me be 100% honest - I know we need to be "good enough" to just do it live but there are a few issues we face. Basically we all met in college but we all live in different states now. The Lead and Tenor live about 1 hour from each other but the rest of us are 10-15 hours apart... SO, we basically NEVER get to see each other or rehearse. So, we get together for a week, spend a day or two singing and making small tweaks to music then spend 3-4 days in the studio... ALL DAY! So, to get 14-18 songs done in that time, we dont have the luxury of making sure everything is perfect. We would LOVE to practice for a few weeks, then go in and nail it in a few takes... and, given that time, we easily could, we're all good enough... but dont ever sing together to make that happen.

Check us out as [[url=http://[/URL]="http://www.NarrowWa…"]Narrow Way - NarrowWayMusic.com[/]="http://www.NarrowWa…"]Narrow Way - NarrowWayMusic.com[/]

I don't know about other pitch control programs, but I believe that if you use some of the methods mentioned above, you can get enough isolation to use Melodyne to correct pitch. How close are the lead and tenor to the studio? Can they go an make some test recordings to see if they can be successfully manipulated? How big is the studio? If you have enough space (8-10' is plenty) I'm sure that singing close to cardioid dynamic mics will give enough isolation. I've done it with full band with drums.

I certainly use Melodyne for pitch correction - GREAT program!!!

If you are using Melodyne I think you can pull this off. Use the techniques mentioned above to try to minimize bleed: Keep the singers as far apart as you can while getting good eye contact and interaction. Use cardiod or hypercardioid mics. Dynamics are better for isolation than condensers. Have the singers get very close to the mics.

Once you have your tracks, plan on spending a lot of time with Melodyne. Experiment a lot with note detection. You don't want the accompanying vocals to be identified as harmonics of the main vocal on the track. Try polyphonic vs. melodic mode. Play with sensitivity. The "Demixing and Remixing a Vocal Clip" video tutorial on the Celemony site talks about isolating a vocal in a polyphonic track.

Yea, I definitely want to get more familiar with that program, still figuring out all the great things it does.

I have found that even minimal amounts of bleed make pitch-correcting a vocal nearly impossible. The main vocal track gets pitch-corrected while the bleed track does not. The dissonance between the two is usually evident to the ear even if the bleed is really low. Your mileage may vary, but I have stopped relying on pitch-correction in these cases.

Jeff

thatjeffguy, post: 382394 wrote: I have found that even minimal amounts of bleed make pitch-correcting a vocal nearly impossible. The main vocal track gets pitch-corrected while the bleed track does not. The dissonance between the two is usually evident to the ear even if the bleed is really low. Your mileage may vary, but I have stopped relying on pitch-correction in these cases.

I thought that too, but for a particular project I was forced to work a bit harder on this in Melodyne and found ways to deal with the problem. In my case it was not that the bleed went uncorrected. It was that the bleed was misinterpreted as harmonics of the "main" note. In general, if Melodyne misidentifies something (most common is to get the wrong octave) the correction has unwanted artifacts. It is important to look for misidentifications before correcting. On an isolated vocal track this involves finding bad octave ID and bad not separations. In a track with bleed it means identifying "mixed" notes. As you say YMMV.

So Bob, are you saying that you were able to isolate just the 'bleed' portion of the audio in a track where it was not the targeted audio? I have and use Melodyne, it seems to me that the identification adjustments eliminate the faintest harmonics first... how exactly are you able to retain the faint bleed signal while eliminating the stronger, intended signal?

Jeff

I corrected the main notes, not the bleed. If you want to identify just the bleed - in note identification mode increase the sensitivity. You can turn identified notes into "potential notes" and potential notes into identified notes by clicking on the individual blobs. You can prevent ranges of notes from being identified with the venitian blind tool. The "Demixing and Remixing a Vocal Clip" video tutorial on the Celemony site has more tips.

Yes, doing what you want is indeed tedious, but its possible. ( I'm not sure why you want just the bleed rather than just the main vocal.) I admit this can't be done all the time. I've given up on some vocal/acoustic guitar tracks with too much bleed. But I've definitely done this with multiple vocalists.

Thanks Bob, I think I failed to communicate my meaning in my previous post. I have no problem going in and pitch-correcting the intended vocal on its intended track. What I'm saying is that by so doing, you put the tuned vocal out-of consonance with the bleed of the same vocal in the bleed track, and this dissonance is usually painfully audible even in minimal bleed situations. I thought you were saying that you could then go into the 'bleed' track and also pitch correct that in order to solve the problem. That's what prompted my question to you. I can only assume that you were dealing with a situation where the bleed was very minimal relative to the intended vocal or I'm sure you would have experienced what I describe.

Jeff