From the breathy smoothness of Diana Krall to the growl of Tom Waits to the screams of Linkin Park’s Chester Bennington, no two singing voices are quite alike. For producers and engineers, the challenge is to find the most flattering way to capture a particular singer’s sound. But successful vocal recording requires more than engineering skill; you also need psychological chops. Coaxing the best performance from a singer is often a lot trickier than selecting the right vocal chain.
There are plenty of different approaches when it comes to recording vocals. While there are areas of general agreement—make the singer comfortable, provide a good cue mix and use a quality vocal chain—when you drill down, you find that producer/engineers all have their own individual techniques, or at least variations of standard ones that they use to capture the best vocal possible.
There’s a lot more you can do before a vocal session than simply setting up a mic and making sure you can hear the cue mix and mic output in the artist’s cans. Whenever possible, learn as much as you can about the singer and the music so you can make the best choices in terms of vocal chain, vibe and psychological approach.
“Preparation to me means I actually visualize myself in the session from the moment I walk in the door until the end of the day,” says engineer Dave Reitzas (Barbra Streisand, Seal, Guns ‘N Roses, Celine Dion and Andrea Bocelli, to name just a few). “I do a mental walkthrough, a training for my mind for what to expect.
“I walk through problems that could arise, like maybe the headphones aren’t right. So I want to make sure that I have another pair of headphones, ready to fix that problem. I want to have option microphones available, I want to have option preamps, option everything—plan B, maybe plan C for anything that could arise. I’ll do a little Internet research and see if there are any kinds of quirks or references or stories, or maybe the artist works with an engineer that I know, and I might give a phone call, and say, ‘Anything I should know about this artist?’ So it’s basically just a little homework to give me an edge in making the session flow seamlessly.”
Foreknowledge of the artist can help you choose the appropriate mic, or at least help winnow the mic choice in advance. “I’ll kind of narrow it down, just by knowing what they sound like based on pre-production or a previous record that I’ve heard,” says New York City–based engineer Joel Hamilton (Tom Waits, Elvis Costello, Pretty Lights, Sparklehorse and more), “or if it’s just the scratch vocal that I’ve heard during basic tracking.”
“More than likely I’ll already have the music if I’m recording the vocal,” points out Miami-based Marcella Araica (Missy Elliot, Joe Jonas, Nelly Furtado, Usher and many others). “I like to get a sense for the feel and tone and what the range is, and that really determines what type of mic I use and what preamp.”
Dave Reitzas says it’s more productive to focus on what a singer is doing well during a session rather than dwelling on what they’re doing wrong.
Knowing which of your mics and pre’s are most flattering for which types of voices and musical styles will give you a big advantage when it comes to choosing the appropriate ones in advance. One thing you want to avoid is spending a lot of time testing out vocal mics and other vocal-chain components while the singer is there.
“My goal is to make the first mic choice the right mic choice,” says Reitzas. “So with my 26 years of experience, I have a sonic imprint in my brain of what hundreds of microphones sound like and which mic would be a good pairing with the singer, especially if it’s someone I’ve worked with before whose sound I’m familiar with. But I basically will pick the mic that I think will work best for the singer and the song, with the intention that the moment they open their mouth, that’s what we’re keeping. There’s so much spontaneity and magic in a first take that you have to have your s**t together to be able to use that.”
Hamilton says he typically has three mics set up before the session starts, but makes his choice quickly. “I don’t even bother with a whole take,” he says. “Usually when somebody steps in front of it and starts talking, it’s as if the glass comes down and you just hear them the way they were just talking to you in the room.”
To give the mics a fair comparison, he uses the same chain for all three mics. “I’ll switch the patch rather than have three mics through three different pre’s,” he says, not wanting to add more variables to the process. “I’d rather just hear the singer. He or she is the variable, and the performance is the variable, and I’m looking for constants at that point.”
He also concurs with Reitzas that it’s best not to spend much time experimenting with different mics at the risk of missing a good take. “You don’t try 50 different cameras right as the bus is about to jump the canyon,” he says. “You go with it, and you do what you can afterward in the editing to tell the story in an engaging way. In general, I would say that I’d go with an inferior microphone choice and a killing take because there are 9 million ways we can make it shine later.”
Araica also sets levels very quickly. “It happens sometimes that they’re in the middle of doing levels and they do that money take,” she says. “I always record. Even if the music stops and they’re in there talking or just messing around, because sometimes they might do something really cool, or their mind might take them somewhere to do a different creative idea, and they’ll realize that they went somewhere, and say, ‘Oh my God, that was so cool, did you record that?’ And I’m like, ‘Yup.’”
CUE IT UP
Another item to check off your list before the session starts is setting up a good cue mix. Obviously, you’ll likely have to tweak it later at the singer’s request, but at least get something in place that sounds good to you and confirm that everything is working.
“I’ll get a great music mix in the control room,” says Reitzas, “using the same headphones that the singer is going to use and the same headphone box, and I’ll put my music on one gr oup fader and then I’ll pull that fader down. Then I go out to the mic and I do a mic check. I’ll get the level on the mic to where I can give it a little bit of volume and make sure that the sonics sound like I expect them to sound. Based on my experience, I’ll set the headphone box to where I think it’s going to be a perfect vocal level for the singer, including reverb and delay levels. I’ll get the vocal to sound perfect without the vocalist being there. Then, while at the mic, I’ll have the assistant in the control room push up the music fader until I think the music is at the right level. So 99 times out of 100, we’re ready to go right from the first note.”
Marcella Araica finds that giving an artist a handheld mic to use, like a Shure SM58, can sometimes do wonders to get a great performance.
GET IN THE ZONE
Everyone agrees that making the singer comfortable goes a long way toward getting a great performance. So part of your pre-session preparation should include learning what kind of environment the singer likes and tailoring the lighting and vibe of your studio accordingly. Low lights and even candles are often used to set the mood. You might even run into the occasional singer whose comfort zone is decidedly non-standard. “Some artists want to sing in pitch black, where they turn all the lights out,” says Araica. “That’s their choice.”
You can also help by keeping the temperature in the vocal booth or live room warm. “You don’t want your vocalist to be freezing,” Araica says. “I think most professional singers understand the importance of keeping their body temperature warm.” She also recommends having room-temperature water available, which will refresh the singer but won’t tighten up their vocal chords as colder drinks can sometimes do.
Producer/engineer Dave Brainard (Jerrod Niemann, Jamey Johnson, Brandy Clark and others) attributes some of his success with vocalists to the relaxed feeling of his Nashville project studio. “It’s got a great vibe,” he says. “We’ve got a coffee machine, and we’ve got wine or beer available. It is a comfort thing. People come up here and they feel comfortable.”
Hamilton contends that the most important factor for bolstering singers’ confidence in the studio is to make them feel that you’ve got things totally in control on the technical side. “It starts with the comfort factor. Just sort of exuding a confidence that I’m grabbing whatever you’re throwing at me in a way that’s flattering,” he says. “Like if you’re sitting next to someone on an airplane who isn’t freaked out by turbulence, you sort of read their body language, and you’re like, ‘Oh cool, we’re not going to die.’ [Laughs] Like this guy has seen this 1,000 times and he’s not freaking out, so I guess I can chill.”
Araica will sometimes use what some might consider an unlikely technique, at least from a studio standpoint: She’ll have the vocalist sing into a handheld Shure SM58. While it can’t compare sonically to expensive studio condensers, the added level of confidence some artists get from holding a stage mic makes all the difference for Araica. “You can get a better performance out of an artist when they’re in the mindset that they’re on that stage or they’re rehearsing the song,” she says. “It’s my favorite mic to use. I can’t use it on everything, but on certain singers I love it, and they know how to open that mic.”
According to Joel Hamilton, you can make a singer feel relaxed by exuding confidence in your ability to capture their performance.
Photo: Jackie Roman
TAMING THE INPUT
Because 24-bit recording doesn’t require as much level to sound good as 16-bit or especially tape, the biggest challenge is keeping the loud peaks in check to prevent distortion that could ruin a take, or part of one. With that in mind, Reitzas, Araica, Brainard and Hamilton all compress to varying degrees on input.
“I use compression when I’m recording,” says Reitzas, “but just subtle: 1 or 2 or 3 dB of compression on the big stuff.”
“When I record vocals,” Araica adds, “I like to really lightly just touch on it, just enough to have some control. You can’t take the life away from it in the beginning process.”
Brainard goes for just slightly more. “The compression I use is just a basic [Empirical Labs] Distressor, and I try to get 3 or 4 dB of reduction.” He tracks through an Endless Analog CLASP system, which means the signal goes through analog tape before hitting Pro Tools. “Now I’m doing all CLASP, and I can’t go back. It’s like night and day as far as what I can do sonically. It just shaves off all those harsh transients.”
Hamilton uses his input compressor, a Neve 33609, mainly for catching peaks. “I wind up with just a little bit, but it’s more like limiting,” he says. “Super-fast release time, super-fast attack time. In the heaviest of sessions, it’s probably pulling back 5 dB in the loudest bit.”
Another way to control vocal-session dynamics is to ride the levels from the control room during the take. “I’m not afraid to ride the level knob on my preamp as we’re recording,” Reitzas says. “Sometimes you get the advantage of knowing the song before it’s recorded, so you know what to expect. An artist will practice in the control room before, and you get an idea of what they’re going to sing. With Streisand, I’m constantly riding the preamp gain. With her, it’s usually quiet verses and big, dramatic choruses.”
Araica adjusts her Tube-Tech CL 1B compressor as needed during the song. “I just use the output,” she says, “or adjust the threshold not to hit as hard. That opens it up even more.”
Hamilton points out that like guitar amps, mic preamps often sound better when set above certain gain levels, and it’s important to set yours to deliver its best tone. “Think about when you’re playing through a Fender Twin,” he says. “Put it on 1 and it sounds like crap. Put it on 3 and it sounds amazing. There’s a point where amplifiers open up and start to show their colors. And that’s absolutely true of a Neve 1073 or a 1066, or a Manley, or whatever you’re using. There’s going to be a point in that amplifier’s gain range where it just starts to rock for vocals. And if you need to, throw something after it [before it hits the converters] to attenuate. I think people would be surprised because they end up with an API turned all the way down and the pad engaged, and somehow their 87 doesn’t have any ‘air.’”
As for EQ’ing on input, Araica prefers not doing so, if at all possible. “You might never see this artist again,” she says. “You’ve got to keep everything safe as possible. In my belief, if it goes in clean, you have nothing to worry about.”
Reitzas typically uses an NTI EQ3 or the new Maag EQ4 in his vocal chain. “The texture I get from those EQs is so important to the vocal sound that I want that quality to be recorded, especially for when I am only called in to record the vocal. This way, the sound is there no matter who takes over on the rest of a project.”
Brainard inserts an API 550B equalizer in his chain. “Usually, I’m just notching a little bit of low end out of the vocal,” Brainard says. “I don’t play with the high end going to tape.” (See sidebar “Chain Reaction.”)
Dave Brainard’s secret for success with singers is to coax them into correcting themselves rather than having it come from him.
“To me the biggest thing in working with singers is more psychological, where you allow singers to self-correct,” says Brainard, “getting a singer to be creative and compelled, and inspired, and [feeling] like they own it, as opposed to reacting to what they think you think is good. That’s the worst place for a singer to be. For me that’s the key to everything.”
Reitzas takes a similar tack. “I stay positive and enthusiastic all the time,” he says. “I am focusing on what I hear that I love. I’m not spending my energy hearing what’s wrong. A lot of artists are very hard on themselves. They’ll do, say, a couple of lines in a verse, miss a few notes and then start bumming because they didn’t get it the way they had hoped for. But I’m listening for the two or three great moments that are definitely ‘keepers’ that sometimes get overlooked.”
When a singer is having problems, especially with pitch, adjusting the headphone mix can go a long way. “I will have the main keys or the piano [on a fader] close to me so that I can push that up in their phones on spots that will help them zero in on the pitch,” says Reitzas. “I am also ready to raise the hi-hat or a loop for help with rhythm. I can’t emphasize enough how important it is to encourage and teach singers who don’t have good mic technique to use the mic to their advantage by getting in close on intimate phrases, and backing off slightly or turning their head slightly on the big dynamic phrases.”
You have to be careful pointing out pitch problems with some singers, who then might end up focusing on it so much that it messes up the rest of their performance. But unless you feel that the pitch issues can be dealt with using post-session tuning, ignoring it probably won’t help either. “It’s like watching somebody running into a wall,” says Hamilton. “I’m not going to tell them to ‘run faster into the brick wall and see if you make it through this time.’ It’s like you can steer them around that roadblock with either technology—changing the headphones, physically changing the environment—or having them come listen to how it’s reading and they usually internalize it right away.”
In the pop/R&B/hip-hop world that Araica works mostly in, it’s not uncommon for artists to monitor through pitch correction in their cans while they’re singing. Although it can introduce some latency into the cue mix, many singers still want it. “There are some artists where that’s the only way they can work,” Araica says.
Vocal tuning technology has made it less imperative for the vocalist to nail every note from a pitch standpoint, at least in some genres. “On modern music where tuning is part of the vibe, I’m totally cool with that,” says Reitzas of vocal tuning. “On true, honest, singer/songwriter music, I don’t like to hear any artifacts.”
Araica, Brainard, Hamilton and Reitzas all say that how you place the vocal mic in relation to the singer varies based on the circumstances of the session. When pressed for their “typical” setup, the four offered some slightly different techniques. “I have a pop stopper about an inch away from the mic,” says Brainard. “I have them sing about two inches away from that, unless there’s a lot of low end, in which case I’ll have them back off another two inches.”
Reitzas and Araica like to start with the mic angled down a little bit, with the singer about four to five inches from the mic. “I usually will put the mic at nose level and then aim it at the mouth,” says Reitzas, who also likes using a very small pop filter to avoid impeding his sightlines with the singer.
Hamilton says he typically prefers an even closer starting point, “so that the capsule is literally like a lollipop in front of the person’s mouth,” he says. “It depends on the shockmount, but as close as the singer can comfortably get, without knocking the thing into the shockmount 100 times with his or her nose—meaning the shockmount is almost touching the pop screen.”
Mike Levine is a New York–area recording musician and music journalist. He’s the former editor of
Reitzas, Brainard, Araica and Hamilton all stressed that the mics and components in their vocal chains depend on the singer and song. That said, they did offer up some of their typical vocal-chain components:
Mics: Brauner VM-1 or Audio-Technica 4060 with Telefunken tubes; mic pre: NTI PreQ3 or Maag PreQ4; EQ: NTI EQ3 or Maag EQ4; compressor: Tube-Tech CL 1B
Mics: Rode Tube Classic, Audio-Technica 4060 or Peluso 2247 LE; mic pre: API 512; EQ: API 550B; compressor: Empirical Labs Distressor; tape-based input hardware: CLASP system
Mics: Neumann Fet 47, U87, U47, GT md 1 or Blue Mouse; mic pre’s: Neve 33114 1084 or 31102; compressor: Neve 33609
Mics: Sony C800, Neumann U67 or U87, Telefunken 251, Shure SM7 or SM58; mic pre’s: Martech MSS-10, Neve 1073 and 1084, Avalon 737; compressor: Tube-Tech CL 1B
Joel Hamilton on Recording Background Vocals
I spoke in depth with Brooklyn-based producer/engineer Joel Hamilton on his techniques for recording background vocals.
Is your approach for backing vocals totally different to lead vocals in terms of setup?
If it’s the same person, meaning like if it’s the singer singing their own harmonies, yes. If it’s a different person in the band, I might just throw them on the mic and have them stand a foot back. I’m big on the idea of perspective, in the sense that you could turn down a vocal all day long, but if you recorded both people an inch away from the microphone, they’d both sound really close to the listener. Conversely, you can turn up a vocal that was recorded 10 feet away from the microphone, and it’s never going to sound as close as the quiet one up close on the microphone. So I do like to position people in space so that you get a consistent image when they’re done.
If you had multiple background parts to record with one singer, would you move him or her around?
Yeah, but subtly. Like put their cheek to the mic but still only—and these are all rough distances—six inches to a foot away, but like singing toward me at the glass instead of straight into the mic, sideways. Or singing toward the back wall even sometimes, just for an ambient take.
What kind of angle?
Forty five degrees, so literally it’s at their ear. Singing across the diaphragm and it’s amazing how different those two will sound when you pan those two off to the sides. Let’s say for like a thickening in the chorus, I usually do it in threes, if I do it at all. If we’re doing doubles, like unison doubles, it’ll be in threes. There’s the main vocal, and one for the people on the left and one for the people on the right of the arena [laughs].
Do you do that with lead vocals?
Again sometimes. Those were the de facto standards in the ’90s rock scenario. I’m not doing it if it’s a girl and a piano. But, yeah, definitely in a big rock record where the dynamics are actually defined by width as opposed to in amplitude terms alone. So meaning when you want the chorus to widen out like crazy because it makes it feel bigger and it wraps around the listener, even though technically it’s not moving the needles any more than in the verse, especially post mastering. So if you want things to widen out, I’d rather get three discrete performances from the singer than just use a stereo delay for the backing vocals.
So the singer would triple it?
Yeah. Sometimes I’ll even use the outtakes. If they’ve sung it a trillion times I won’t even bother having them do a specific one. I’ll use the outtakes as the doubles in the chorus, let’s say, and maybe even pitch-correct those a little again if it’s appropriate. Sometimes I even use pitch correction to create a double, and even draw it in—actually draw in the double so that it feels like an exact rhythmic unisonbut with slightly different pitch. Not even because it was better or worse, just to move the pitch around a little bit.
If you have a group background vocal, would you typically use one mic in omni?
If they’re singing at the same time, I’ll probably do something like figure-8, separating two groups of three with low and high or something. I just did a record with a group called the Parkington Sisters. And we ended up with two figure-8s in a Blumlein configuration. And the way those vocals sum in the air is just something that you could never do with five faders. It’s incredible. It gave me the chills every single time, the way those two capsules were being engaged. Some of them were single microphones, but never in omni. I rarely do that. I will put it in figure-8 and group people shoulder to shoulder, facing each other like three on three or two on three. That way, I can back up the melody side, and I can get my distances and put a piece of tape on the floor that’s like for the two higher voices and then the three that are doing supporting.
However the harmony works, you can usually compartmentalize it into melody and support. And I’ll do that because it’s easier, even semantically in the headphones, to say, “bass side back up” and we’ll get the balance that way.
In a more generic situation, where it’s not sisters, would you still do the same thing with group vocals?
It depends how they sing together. If it’s four dudes doing barroom vocals, yes, if there’s really distinct intervals where we need to concentrate on each one to make sure that nobody’s poisoning the well [laughs]. Because sometimes you get four people in there and you can’t work on the middle interval ever again. If they’re killing it on one mic, sure, but if they’re not, then it’s an individual divide-and-conquer situation.
Why not omni in a group vocal situation? It sounds good in theory.
That’s because everyone has seen the drawing of a circle around the microphone, and you think of omni as being everywhere, but it sounds like crap. A trash can is a circle, too. You can’t rely on the response sounding flattering for a vocal all the way around. I just think there’s a lot of misconceptions about omni. If you put a 47 in omni and ask somebody to sing really close to the microphone, it sounds surprisingly cardioid, only worse. And with certain microphones, like my Soundelux U95, I can put that in omni and have somebody sing really close and it sounds gorgeous. But it’s not just a guarantee that you’ll get what you had in cardioid mode, that you’ll get that somehow in a 360-degree [field], like a magic cardioid microphone that’s aiming at 360 people’s mouths in a circle. It just simply isn’t true.