Neumann KH 80 DSP 160×600
REC Mixing Workshop 160×600
Adam Audio T-Series 728 x90

A well-prepared vocalist plus a well-chosen mic equals a perfect recording

By Paul Stamler

The human voice—the universal constant. All cultures use it for music, for storytelling, for oratory. Why is it so tough to record?

The answer to the question at the top of this page is in the question: it’s tough because it’s universal.

We’ve used our voices for thousands of years, perhaps hundreds of thousands—anthropologists are still arguing over when humans began to speak. But they agree that over many millennia we’ve evolved acute sensitivity to the nuances of vocal sounds. To survive as social animals, people became closely attuned to the sounds their neighbors were making, using the minutiae of vocal inflection to recognize each other, to pick up meaning, to sense the emotional states of their fellows. The ones who failed probably got hit over the head.

So we’re good listeners, and we’re especially sensitive to the sound of the human voice. It’s a complex sound created by a double-reed system and amplified by a variety of resonant chambers in the chest, mouth, and sinuses. The variety of possible sounds is immense (listen to a recording of a good “human beat box” or that style’s ancestor, the C&W “eephing” vocal style, sometime), the radiation pattern complex, and the technical problems daunting.

But we manage, and in this article I’ll make a few suggestions for getting the most out of a singer.


The pipes

Please note: I’m not a professional voice teacher, just a singer and recording engineer; the ideas I discuss are nuts-and-bolts suggestions I’ve picked up over a few decades, but they’re only snippets. I’ve yet to find a good book or training course for singers like me; most of the material out there is aimed solely at classical singers. We pop, rock, and folk types tend to be shortchanged. Still, I’ve learned a few things, and I’ll try and share ’em.

The first stage, as with any musical recording, is making sure the instrument is in shape. Think of Mick Jagger: before a Rolling Stones tour, he works out in the gym for several months. Part of this, of course, is to build up his strength for cavorting across the stage, but it also serves to improve his wind and build up his lungs, where the voice begins. Now, I’m not suggesting you run out and sign up for a gym membership, but a few weeks of light jogging before an important recording project can work wonders on your vocal stamina.

Next, think of the vocal production mechanism itself. The human voice is made by the actions of muscles, and if you neglect a muscle it gets flabby. If you’re gigging regularly you’re probably in okay shape, but try doing a little extra singing in off-hours. You’ll be surprised how strong your voice can become; after six months of singing sea shanteys in a noisy pub, sans PA, my voice was powerful enough to wake the dead. (Some of the folks in that pub were good candidates.)

It didn’t hurt that every couple of weeks I’d get together with a dozen like-minded sorts and sing shape-note hymns, a traditional American art form notable for being sung at only one volume: too loud. (It also didn’t hurt to have the right genes for this; my old man worked his way through school selling hot dogs at ball games, and the family name means “town crier.” We’re a loud crowd.)

Speaking of your voicebox, keep it warm. If it’s cold out, wear a scarf around your throat to avoid chilling those muscles.

Kef America 55 Years of British Engineering – 728×90

Get up, stand up

Your lung capacity is greatest when your body is in a straight line. Don’t believe me? Try singing a sustained note at constant volume while sitting down with one leg crossed over the other, then try it while standing. Trying to sing while seated is an uphill battle, as you’re likely to run out of breath at the end of lines or phrases unless you power down your singing. The object is to maintain a constant air flow rather than a series of grunts or gasps, and if you trail off at the end of each bit because you’ve run out, you’ll sound like a wimp. That constant flow is the most important thing I’ve learned about singing. Spending time imitating a bagpipe is good practice; it’ll improve the power of your singing by an order of magnitude, and your neighbors will be so glad when you stop.

Anyway, get your body in a straight line by standing up in front of the microphone. If you simply can’t work standing, try sitting on a (non-squeaky) stool; it lets your torso remain straighter than most chairs.

Now check your head. It should be level and straight—not a comment about your attitude or orientation, but a suggested position. Try this experiment: hold yourself with your head high and looking straight ahead while making a droning “ah” like a bagpipe. Hunch your head forward and slightly downward; note that your jaw can’t open as far as before. Bad idea.

Now throw your head back while continuing to drone. Feel the tightness? This is another bad idea; if you look at old photographs of folk singer Pete Seeger, you’ll find he spent a lot of time singing in that position. Expressive and energetic—but his voice is now almost entirely gone, after years of unknowing abuse.

Incidentally, some people find they gain a bit of extra vocal power from the slight pressure of a guitar’s body against their abdomen. Even in an overdubbing session, you singer-guitarists might try wearing your guitar and seeing if it helps. I also find it useful to hold some vertical object—my guitar neck, a mic stand—when singing a cappella. This is probably more psychological than physiological, but if it works….

Speaking of psychological tools, there are many mental flip-flops that singers use to improve their sound. One that’s useful to me is to compose a visual scene in my head that conjures up the mood of the song; it can be something representational or something abstract. I also like to imagine that the sound is coming from a small ball about an inch across hovering directly in front of my mouth, then use my muscles to place it there.

Some singers like to imagine they are singing the song to one particular person, and visualize that person’s reactions to each line or verse. Others pretend (or persuade themselves) that the song is singing itself, and they’re just along for the ride. There are probably as many varieties of mental jiu-jitsu as there are singers; it’s a subject worthy of its own article.


Aunt Jane

About 20 years ago a fine singer named Jane Grosby (now Jane Vidrine, half of the Magnolia Sisters) gave me a list of items that affect a singer’s throat for better or worse. It’s been a long haul, but here is what I remember of her suggestions; additions and errors are of course my own.

In general, thin liquids are better than thick. Juices are better than milk products, which leave a gucky film in the throat. Almost anything containing citrus is good; I’ve had great success myself with Celestial Seasonings Lemon Zinger herbal tea or the equivalent, especially dosed with honey. [Music Maker Publications does not endorse any particular product in this vein, despite the fact that our offices are right down the street from Celestial Seasonings’ world HQ.—Ed.]

Coffee, cola, or anything else containing caffeine is bad news; caffeine badly constricts the throat. (This always leaves me in a quandary; I play much better after an espresso or two, but my singing goes to pieces. Sigh; life is full of hard choices.) Real (non-herbal) tea is even worse; on top of the caffeine it contains tannin, which is also constrictive.

Many musicians like to loosen up a bit with alcohol before recording. If you’re one of these, do it in moderation; tapes by drunks tend to end up on blooper reels. Alcohol does curious things to the throat. White wine opens it, red wine closes it. (Tannin again.) Scotch whisky closes, bourbon whiskey opens. Vodka opens. American-style beer seems to be pretty neutral, but stout is too thick and syrupy for the best singing. I have no information on gin, rum, or brandies. Whatever you do, don’t use booze to kill the pain when your voice is strained and hoarse; without the pain warnings you may get through a session at the cost of permanent voice damage.

Several professional singers of my acquaintance recommend Ricola herbal cough drops to soothe and lubricate dry membranes. I used them for a while, but they work a little too well; they stimulate the salivary glands to the point that singers can sound slurpy. (They’re great off-duty, though—in fact, one is lodged in my cheek as I write.) Less fancy cough drops are less spitty, but also less effective.

In the end I’ve settled into the simplest throat treatment of all, on the suggestion of shape-note regular Bob Borcherding: plain hot water with perhaps a twist of lemon. In the long run it’s proved the best compromise, and it won’t give you hangovers or leave your fingers sticky.


No ifs, ands, or butts

Even if you’re a confirmed smoker, you should make the vocal recording area of your studio a permanent non-smoking zone. The built-up tar residue from years of cigarettes acts as an instant throat constrictor to many singers, even some who smoke like chimneys themselves. The diaphragms of your microphones will thank you too, as will your back-up vocalists.

Heck, you might even consider quitting altogether. You’ll live longer—and giving up cigs for a few extra decades of sex sounds like a pretty good tradeoff to me…but I digress.


Studio time

Okay, we’ve got your voice into the best possible shape. Let’s put up a mic and start recording. The classic position for a vocal mic is 6-10″ from the singer’s mouth, straight ahead. Most of the time this gives a nice, balanced sound; the singer’s slight movements don’t affect volume levels drastically, tongue and saliva noises aren’t too prominent, and most good vocal mics are tailored for flat frequency response at a distance of 6″ with the bass roll-off switched on.

A lot of performers, however, prefer to get closer. Much of this comes from live performance experience, where it’s necessary to be right on top of the mic to avoid feedback at elevated volumes. Others simply prefer the extreme “up-close and personal” sound that comes with close mic placement. (I confess to being a confirmed mic-eater; using dynamic mics, I usually stand with my moustache just brushing the windscreen.)

When using a very close mic, especially on a man’s voice, you’ll probably need heavy bass rolloff; the “proximity effect” in cardioid microphones can boost bass up to 10 dB at close distances. (Of course, this may be useful in beefing up a thin voice or adding “chest-tone.”) Close mics also suffer from severe volume changes when the singer moves a small distance; if your vocalist can’t stay put, you may need to use serious compression.

A useful trick to back enthusiastic mic eaters off a few inches is to use a pop filter. Another, truly sneaky, is to set up two microphones—one close in, one 6″ away. Mute the close one. Surprisingly enough, this trick even works when you try it on yourself.

Wherever your microphone ends up, put pieces of gaffer’s tape on the floor in front of the singer’s toes so s/he can return to the same spot for subsequent takes.


Back up and cross

So far I’ve mostly addressed the problem of recording lead vocalists close up with a single mic—the usual situation in pop music recording. However, it’s worth noting that backup vocals sometimes call for specialized techniques. Our resource library has some great articles specifically on backing vocal production, but here are some basics.

When there is a group of backup singers, especially if they’re used to working together, I usually prefer to record them grouped around a crossed pair of microphones at a distance between 1 and 3 feet, depending on the singers, the room, and the degree of intimacy desired.

By setting up the microphones in an XY pair (diaphragms stacked, angled at 90–110 degrees, panned hard left and right), the singers are captured in true stereo. A stereo image of the acoustic space is created, one that sounds more real than any artificially generated space I’ve heard. The singers control their own vocal blend rather than having it set by your itchy fingers in the mix. This method isn’t for control freaks since it means you have to trust your musicians, but it sure sounds sweet.

In fact, the technique isn’t only for backup singers. Many times I’ve recorded lead vocalists with crossed pairs, often as close as 4″ from their mouths, and gotten richer, more solid sounds. And one time I recorded an a cappella gospel quartet with a crossed pair—of ribbon microphones, no less—and was rewarded with a recording that is spookily real-sounding.

Antelope – Edge goes native

Peter Piper Popped a Pile of Pesky Plosives

Popping “P”s are a perennial problem in vocal recordings—and not only when close-miking; I’ve had an operatic soprano blast a KM84 at 15 feet. The letter “P” (and to a lesser extent, “B,” “T,” and “D”) can produce a short puff of air that is the equivalent of an extremely loud sound at a frequency too low for our ears to hear.

But our microphones hear it to varying extents. Dynamic mics usually make a mild “thud” as the diaphragm is knocked out of its linear region. In condenser mics the breath blast produces a large pulse of electric charge from the capsule, which overloads the amplifier circuit in the head of the microphone. Some circuits take up to half a second to recover from P-pops; during that interval the amplifier is blocked so the rest of the vocal sound drops out. The net effect is ghastly.

The first line of defense is in the mic itself; most microphones have built-in mesh and/or metal screens that break up breath blasts with varying degrees of effectiveness. Most of the time they need help, often from a foam windscreen that slides over the capsule. Curiously, pop filters (foam or mesh) sometimes work better up-close than at medium distances. Several times a filter that completely stopped the pops from a mic-eater was less effective when the mic was 6″ from the singer’s mouth—the “ideal” position.

The design of pop filters is a delicate balancing act; the thicker the foam, the more effective it will be against pops but the more likely it will be to absorb high frequencies as well, which you usually don’t want. (A few microphones, such as the Neumann KM84, actually have smoother high frequency response with their pop filters on, presumably due to the forethought of their designers.)

For studio use, the trade-off isn’t important; you’re better off using an external pop filter anyway. Commercially available pop filters exist, usually selling for $20-30; as an old-school guy from before the era of easily available models, I’ve always made my own using a panty-hose and embroidery-hoop design I learned from George Vazquez.

Panty hose and embroidery hoops

Start with a common embroidery hoop, available from a sewing or notions store—possibly a drug store. There are several sizes and types; I find a 6″ hoop ideal for most recording work. The ones with springs on the outer ring rather than clamps are the easiest to mount.

Now cut the toe sections from a pair of Sheer Energy panty hose. (Heavier hose absorbs high frequencies.) Disassemble the hoop and stretch a toe section over the inner ring until it’s tight. (Hold it up to your ear to verify its acoustic transparency.) Hold the hose taut with one hand while you twist the ring several times, making a tiny twisted neck on one side of the assembly.

Now, still holding everything taut, slip on the outer ring and let it spring together, clamping the panty-hose in place. Make sure the twisted bit isn’t right next to the spring. Now cut off the excess hose, leaving about 1/4″ of the twisted bit hanging.

If you’re using a spring-type hoop, you can attach it to a long, flat metal rod (about 3/16″ wide and 12″ long) by slipping the rod under the spring and sliding the hoop about 1/2″ onto the bar. The rod can then be lashed to an unused microphone boom or gooseneck using gaffer’s tape or rubber bands, and positioned between singer and microphone for best effect.

I’ve found it works best when the pop filter is about 4″ from the mic and the singer’s mouth is 2–4″ from the pop filter. If you don’t have a metal rod, a thin dowel rod or chopstick works fine.

In a pinch, working with dynamic mics, I’ve clamped the back of the rod to the mic’s own boom, letting the middle of the rod sit on top of the microphone’s ball end or foam screen. If you position everything carefully, the filter will sit a few inches in front of the capsule and the angle of the rod will center it in front of the mic.

My first experience with this filter came when I was recording Phil Cooper and Margaret Nelson sometime around 1985. I was having problems with P-popping on the beyerdynamic and Neumann mics, and I remembered the filters my friend George Vazquez had described to me. Phil’s mother supplied the hoops, and after an unsuccessful attempt using Margaret’s stockings (too heavy), Phil and I headed out to find some Sheer Energy.

These were the early days of digital recording; our primary recorder was still analog (my faithful old Revox A-700), but we were using a Sony PCM system for back-up. This encoded the digits onto an NTSC video signal, which we recorded on a standard home VCR. Tape was running low, so we added that to the shopping list.

Picture the scene: midnight at an all-night drug store in South Elgin, Illinois. Enter Phil and me—two bearded characters looking like Henry VIII and a short truck driver, walking up to the checkout counter with four pairs of Sheer Energy panty hose and three blank videotapes. I think we made our reputation in that town for a long time to come. But I digress…

Microphones big and little

The classic microphones for pop vocal recording are large-diaphragm condenser mics, most famously the Neumann U87, U47, and M49.

I’ve had no experience with the latter two (the prices took off into the stratosphere years ago), but I used the U87 for many years. It has that elusive quality of “presence”; the sound grabs you and makes you listen in a way few microphones can match. Its broad, gentle treble hump adds intelligibility in a way I can’t quite match with any eq I’ve played with, and the liquidity of its tones are often the best choice for many vocalists, male and female.

The U87 has less proximity effect than would be expected in a cardioid microphone, so it is less likely to require bass roll-off in recording—and it’s also less prone to change response with slight changes in the singer’s position.

Because the U87 is multi-pattern, it can be used as an omnidirectional mic to increase the pick-up of room ambience (and further decrease proximity effect), or used in figure-8 mode to record two vocals at once (as can the original U47, as in the Beatles’ “A Hard Day’s Night”).

Other great condenser mics for vocal uses include the AKG C12 and its successors the C12A and C12VR, and the Sony C37. All share the U87’s sense of presence and its flexibility. The problem, of course, is cost; at many thousands of dollars each, these classic condensers are way beyond most people’s budgets, including mine. Fortunately there are hundreds of mics available at more reasonable prices nowadays, including large-diaphragm mics primarily intended for vocal recording.

A perennial favorite has been various editions of the AKG C414, with a clear sound that works on some singers, not on others. (One friend of mine with a difficult sibilance problem found that the C414 was the only condenser mic she could use for recording.)

Another condenser mic I’ve found useful is the Shure SM81. It can be a nightmare to use on vocals, as its response to P-pops is easily the worst I’ve encountered, but it can be tamed using the built-in 10 dB pad and steep bass cutoff plus a pop filter and a foam windscreen. Why bother? Because for some vocalists it’s one of the cleanest, most natural sounding mics available. (One of those vocalists is me.) Sibilants can be a little steely, but the midrange tone is sometimes quite convincing.


As I’ve written elsewhere, I don’t think dynamic mics get a fair shake among recording engineers. When thinking vocals, we automatically think “condenser,” but I’ve had excellent (often superior) results with good dynamic and ribbon mics. I could spend many pages discussing them, but let me cover the ground briefly.

My favorite dynamic mics have essentially flat responses. For men I usually choose the Electro-Voice RE15; its lack of coloration and minimal proximity effect provide natural response at almost any distance. For a smoother, creamier sound on both men and women (so creamy I worry about my cholesterol), the Shure SM7 (set with the treble flat, bass rolled off) does a nice job.

When I want a slightly brighter sound, I’ll go for an E-V RE16 or RE20, with a slight hump in the mid-treble—just enough to cut through a dense mix while still sounding natural. Brighter still are the beyerdynamic M88 (my favorite hot rock-and-roll-type microphone) and the Sennheiser MD421.

For women, my first choice is always the beyerdynamic M260, a ribbon mic that combines lovely warmth in the midrange with great detail at the top. Its pronounced proximity effect puts extra oomph into the chest tones, adding richness to voices that can otherwise sound a little thin. (I’ve even found this useful on a couple of men’s voices, most notably a blues singer who sounded squawky on more conventional mics.) And an old RCA ribbon (or the modern equivalent from AEA) can warm up a lady’s voice like nothing else.


Bits and pieces

Always use a shockmount, even when recording vocals by themselves. You’d be surprised how much foot tapping sneaks its way up the mic stand and onto your recordings. Condenser mics in particular can pick up the subsonic signal and create distortion…. in rhythm.

It’s standard practice to mix vocals with lots of reverb. Just for grins, some day try them with none at all for a very different, very “in-your-face” sound. (See The Band, ca. 1969.) Or try mixing the voice with only slapback (à la Sun Records, 1956). No one said you have to sound like everyone else.

Time to go try some more microphones. Keep singing!

Kef America 55 Years of British Engineering – 728×90
JBL 3 Series MkII (ROS)