Recording a group of singers at home…
By Linda Taylor
My film-composing partner and I needed to record some score ideas for an upcoming feature, a Disney-esque family film. We wanted an organic Americana approach—churchy, intimate, and acoustic.
We wrote the tune and arranged it for a small a cappella choir. The session was recorded and mixed in my project studio.
As the title of this magazine says, I am a recording musician, not a recording engineer. My recording education and experience has been through trial and error, studying, and asking a lot of questions. That being said, I like to do practice runs before any session. Set up mics, experiment with placement, check all the headphones, cables, etc. If I appear confident and relaxed, the performers will take their cue from me. I’m very big on pre-session planning.
First thing was to arrange the physical tracking room. My studio is in a long converted family room. On the plus side, it has high ceilings and carpeting throughout. On the minus side, French doors take up nearly one length-wise wall. Perpendicular to that is an enormous brick fireplace, useful for, say, roasting a wild boar.
I wanted a dry recording, small, warm and intimate, as little room noise as possible, and all the glass, brick and boar space wasn’t going to make it easy. Using Auralex baffles, I lined one wall, and hung heavy curtains across the French doors. Creative sofa placement and lots of overstuffed pillows took care of the enormo-fireplace (besides giving the artists somewhere to sit).
Acoustic treatment stuff out of the way, it was time to take stock of the gear list.
My usual signal chain is Avalon VT-737SP preamps through a MOTU 1224 interface into Digital Performer. This path provides me with plenty of low end and depth, and is complemented nicely by condenser mics. Although I briefly considered using ribbon mics on the choir, I knew tracking time was going to be tight, and I didn’t want to subject the choir to a lot of experimenting during the session. So I chose two Neumann U87s—a studio workhorse with a familiar sound.
Microphones selected, the night before the session I started experimenting with placement. Consulting my well-worn copy of Professional Microphone Techniques and several online articles, I narrowed it down to two choices: Spaced microphone placement (two mics set symmetrically in front and directly facing the singers), and Coincident (XY) microphone technique (a matched pair of directional mics placed as closely as possible to each other, right in front of the center of the vocal group, with the diaphragms of the mics angled about 90-135 degrees).
Both methods afford a natural stereo spread, but the spaced pair has a potential for problems: comb-filtering, lack of mono compatibility, vague center imaging (which was my biggest concern). I decided that the XY method would be my first choice, and the spaced pair my backup plan.
I set the Neumanns to the cardioid position, hung one mic upside down, and the other upright, directly underneath. The mics were so close that they were nearly touching, so I felt safe from phase problems, and their cardioid patterns rejected the noise from the computers and rest of the room. While recording, I walked back and forth in front of the mics, clapping my hands, singing the National Anthem and other favorites, and moving in to test for proximity effect. The playback demonstrated the XY technique as highly accurate with excellent imaging, a well-focused center, and virtually no phase problems.
Prep work done, bring on the singers…
The session—settling in
Five vocalists made up this small choir: bass, tenor, two altos (one singing the lead) and soprano. The lead alto has a warm, earthy sound, almost country without the twang. The soprano and bass were both able to control their vibrato and bravado nicely, blending well with our lead. Even though they didn’t have a ton of recording experience, they were a solid, well-rehearsed group.
The singers stood in a tight semicircle, surrounding the mics. Lead alto in the middle, flanked by the bass and soprano on one side, tenor and 2nd alto on the other. While my partner warmed them up, I took the opportunity to stand in front of the singers, and moved them back and forth until the blend was right.
I had some brief moments of panic; one, the bass vocalist is quite tall—well over six feet—while the lead alto is more on the tiny side. I played with the mics trying to find a happy medium, and figured it would be best to have him sing straight on, and for our lead to raise her head and aim up towards the mics. My other panic attack: the lead alto has a high end that is very salty, and I wondered if the Neumanns might be too bright on her.
Compression and placement
I set up a couple of pairs of mono tracks in Digital Performer. I used mono pairs instead of stereo tracks, because I wanted some control with left/right spread, if even only a little, at mix time. My usual custom is to record with a slight bit of compression from the Avalons, about a 2:1 ratio with just a couple of dB of reduction, just enough to get the level a little hotter to “tape.”
In this case, it wasn’t that easy. Our arrangement called for the lead to sing the first verse solo, followed by the choir’s entrance in the second verse. Once the choir kicked in, I was getting way too much slam from the compressor. I played with several settings with little success, and started moving the singers around again. Our lead wound up standing about two feet in front of the rest of the choir, and about a foot off the mics. Although this ruined the choir’s sight line, it helped the recording, and my initial compression settings were restored.
First rule of thumb when tracking singers: Record Everything. Second rule of thumb: No matter how many times you ask them to seal off the unused cup of the headphones, they will not.
I recorded several takes of the warm-ups, then we listened to playbacks and made final adjustments. As I suspected, we were getting bleed from the headphones, worsened by the fact that our lead alto would take hers off once she heard the opening pitch. My frequent reminders to the choir were not enough, so I needed to mute and un-mute headphones constantly during recording.
Tracking went smoothly from that point on, and after a couple of hours we got the “keeper” take along with some overdubs. We sent everyone on a break and did some critical listening. There were unexplained pops and clicks on the tracks, which could have been the mics clanking together as the singers moved. But it sounded digital to me—there’s really no mistaking that noise. Though I would have loved another take or two, I could tell the choir was finished for the day, so this would have to be “fixed in the mix.”
Okay, bring on the mix….
I needed to address the pops and clicks, later traced to buffer problems within my Digital Performer environment (since then fixed—see upcoming DP 3.1 article). Though I hid these artifacts from the choir during session playbacks, during critical listening the noise was very noticeable.
Using the new Restoration Bundle from Waves, I was able to rid the tracks of virtually all the clicks, with little effect on the original audio. This is an excellent, albeit pricey, tool for exactly this type of audio problem. Best part: the manual lists the sample processing delay, so you can easily adjust your tracks accordingly.
The next issue was the headphone bleed: not too bad on the main tracks, but really bad on the overdubs. On this exposed arrangement the bleed was quite audible, almost phase canceling the lead vocal. Edge editing of the overdub sound bites took care of most of the problem, but only by cutting very close to the audio. This effectively erased the backgrounds’ inhales, but I still had enough ambience from the lead vocal and main choir tracks.
On to the fun stuff. The Neumanns performed wonderfully, giving that chesty, breathy, presence thing in the higher registers. Our lead alto didn’t sound too bright at all; the U87s brought out her high end in great detail, and I love the way they put a sparkle on all the voices.
Plug-ins and processing
The main left and right tracks were bussed to an Aux track for a little processing. Using the Renaissance Compressor plug-in (also from Waves), I applied a touch of squish, just trying to put our lead back in with the choir a bit. When the lead was soloing there was barely a tickle of gain reduction, while there was no more than 6 dB total reduction during full choir peaks.
Setting the comp took quite a few tries. At optimal settings it blended the lead into the choir perfectly, but brought forward the tenor too much. Concessions were made. I used the same plug on the overdubs’ Aux, but with a lower threshold and much higher ratio, and more constant compression ranging from 2 dB to 6 dB of reduction.
Equalization was very straightforward. On both the main tracks and overdubs I needed to roll out a little tubbiness around 300 Hz. But to keep the warmth I chose to cut it a little higher, around 350–360 Hz. I couldn’t resist a little high-end stuff, so all tracks got a little hype around 10–15 kHz.
Of course, once the compression and hype was added, I needed to adjust the inhales on the lead alto, which were now overwhelming. The automation features of Digital Performer make this operation painless. I recorded the volume automation once through manually, then went into the Sequence Editor and fine-tuned the volume curve.
I used some hall reverbs from a Lexicon MPX 500. All tracks had the same size ‘room’ but I gave the overdubs a touch more pre-delay and reflection for a sense of depth.
Before I complete a mix I like to bus the tracks through a TC Electronic Finalizer Plus. Starting with the “Neutral” preset I disable all processing except Compression and Limiting. I set the compression minimally, using a ratio of about 1.8:1, and the threshold for about 4 dB of reduction. Monitoring through these final settings highlights any remaining problem areas.
In this case, only slight volume adjustments were needed on the overdub tracks. Mix completed, I recorded the mix through the Finalizer back to Digital Performer, did the dither two-step, and started burning CDs.
This project was a perfect example of the value of pre-planning and good mic placement. As a matter of fact, this particular session was exclusively about mic placement—once it was time to mix, there was actually very little to do.
It also brought home the importance of properly casting a session. Our lead alto did a wonderful job with the song, her voice was exactly the flavor we needed, and the choir had the skill to follow her reading.
Linda Taylor is a guitarist, composer, and music producer in the Los Angeles area and can be seen on ABC’s “Whose Line Is It Anyway?”. Linda may be contacted at firstname.lastname@example.org, and you can check out her website at www.rubiconmusic.com.