© John Klepko 2013

RECORDING VOCALS

1. dampen the acoustics

2. a music stand can present sonic problems

3. “popping”

4. sibilance

5. most close-mic’d vocals are compressed

6. proximity-effect (bass boost at close distances)

7. vocalist that tends to move around a lot

8. towards the production end of

recording vocals

9. vocalist in the control room

10. importance of a really great headphone mix

11. if you want to avoid using headphones

12. “comping”
© John Klepko 2013

1. Unless trying for a specific room sound effect, it is usually best to deaden the acoustics of the
room around the vocalist – otherwise certain notes/words will tend to excite the room reflections
more than others giving an unbalanced effect- or, the room sound will impose itself on the vocal
sound – making it difficult to add any other effects in the mixdown stage without the room
character getting in the way. And, consider that these reflections are picked up in mono, panned
center along with the vocal mic track, and made more apparent by the compression that is often
applied.

To deaden the room acoustics; put a carpet on the floor, and place sound absorbing baffles
around the vocalist - a baffle/gobo behind the singer is the most effective placement if you have
only one available.

1
© John Klepko 2013

2. A music stand can present sonic problems from strong reflections that cause comb-filtering. This
is more noticeable when the singer moves around causing a “phasey” sound quality. The larger
solid metal stands can even resonate (or ring) at certain pitches. Solutions are to place some
sound absorbing foam material (or cutout piece of carpet) on the stand to dampen both the
reflections and the ringing tones. Also, try to orient the angle of the stand to redirect reflections
away from the vocal microphone as much as possible.

3. Close mic’ing (within 8”) of a singer can lead to problems with “popping” caused by the plosive
consonants (such as b, p, t and sometimes ‘h’). These consonants produce a strong focussed wind
turbulence that vigorously upsets the microphone diaphragm resulting in a distortion that is low-
frequency heavy, giving a deep “thump” or “pop” sound. These pop blasts are increasingly a

2
© John Klepko 2013

nuisance since more listener home systems now add a subwoofer. This is mainly a problem with
condensers and ribbon microphones due to their light diaphragm element and natural low-
frequency resonance. Dynamic moving-coil mics (like a Shure SM-58) specially designed for close
vocals have a spherical-shaped “pop filter” attachment that further reduces and disperses the air
flow energy that causes pops. Still, most vocal recording situations have the need for the type of
detail and range provided by a condenser mic. Some possible remedies for popping are:

• place the mic above or below the singers mouth avoiding the direction of air flow…this may
not always sound the best, as well, the singer often can’t help but sing towards the
microphone (wherever you place it). Placing it above the mouth results in a more pinched nasal
timbre.

• have the singer quickly turn away whenever they sing a plosive consonant….but, the singer has
enough things to be concerned with trying to achieve a good performance without worrying
about such technical issues that have little to do with artistic expression…but some singers
can have the presence and control of mind to do this (…for you).

• use a foam pop filter…some don’t work as well at reducing pops, and they all do reduce the
high-frequency energy (through absorption by the porous foam material).

• use a fabric pop filter. Sometimes called “popper stoppers”. For real close mic’ing, they still
don’t get rid of the pops completely. They can be a nuisance to set up and also annoying for
the singer to have this odd thing in front of their face. But, they are the most commonly
found solution . They can be bought for about 25-50 dollars, or more cheaply, made from
nylon stocking material stretched over a loop made from a knitting loop or bent wire coat
hanger.

• use a low-cut filter (hpf) to reduce the apparent pop sound at low frequencies….this can rid of
most of the annoying sound, but there are still many overtones that come through. It's really
just a quick & easy band-aid solution since the microphone’s diaphragm itself over-reacted to
the wind blast.

• use a limiter with a low-frequency boosted version of the vocal track inserted into the
sidechain…only instances of severe low-frequency energy will trigger the limiter action. Again,
this is a solution only after the pop blast has occurred at the diaphragm.

• an old trick is to gaffer-tape a pencil vertically across the microphone capsule. This helps
disperse some of the wind blast.

• use a pressure-type omnidirectional microphone (instead of the more common cardioid

pressure-gradient mic). The lack of proximity-effect and the typically much tighter
diaphragm tensioning causes less bass boost, and less shock movement of the diaphragm.

4. Sibilance is a problem similar to “popping”, but in the high frequency range. This is the result of
“ss” and “sssch” vocal sounds that cause a focussed air flow on the diaphragm….but most of the
problem is due to erratic (non-flat) frequency response in the higher frequency range (between 3
kHz and 12 kHz) resulting in a harsh, whistling distorted hissy sound at the occurrence of any “ss”
sounds. Certain microphones would work better with certain singers in this respect…but that may
change on another song, or another day, so you can’t always be sure that a singer/mic combination
will always work….The act of compressing vocals themselves can often exaggerate sibilance

3
© John Klepko 2013

distortions since the detector circuit won’t register the “ss” sounds as being as high in level as
the vowel sounds (and other consonants), and the result being that compressor gain-reduction may
apply to anything but the sibilant sounds – leaving those to be higher in relative level at the output
of the compressor than they were at input. The most common solution is to use a specialized type
of compressor called a “de-esser”, or a regular compressor (with a side-chain input capability) can
be converted into a de-esser by inserting an EQ’d version of the vocal track into the side-chain
trigger input. The EQ would be set to boost usually somewhere around 4 – 6 kHz (but sometimes
beyond that range) and cut off lower frequencies below 1 kHz.

5. Most close-mic’d vocals are compressed. Because it is usually the most prominent element in a mix,
care should be taken to have the compressor 'sound' transparent while still being effective.
Usually the compressor should be set to gain-reduce by about 6 - 8dB on the most loud parts of a
line (or word). This depends on the input level, but the ratio is set rather mild between 2:1 and
4:1. Start with the attack and release at a medium-fast setting and adjust according to the vocal
phrasing. Too fast an attack will have the effect of “punching holes” in the start of the words; too
fast a release may result in “pumping” artifacts; too slow a release will bring subsequent syllables
and words down in level when they don’t need to be. Often, compressors with an “auto”
attack/release setting is good since this is in effect, a variable setting dependent upon the nature
(and level) of the input signal….so-called “soft-knee” compressors are also good to have more
gradual compression (via increasing ratio) near the threshold level.

6. Proximity-effect (bass boost at close distances) is a big part of many singer’s “sound” and should
be used to an advantage to make a vocalist sound big, warm and relatively loud (remember the
Fletcher-Munson curves). But some microphones have a more pronounced (or sensitive) proximity-
effect. This can create problems with an inconsistent low-end with only slight movements of the
performer…the degree of inconsistency is worse the closer the general working range of the
singer is to the microphone….it can sound like a variable low frequency EQ.

7. Microphones with tighter polar patterns can be a problem with a vocalist that tends to move
around a lot. Figure-8’s (such as most ribbon mics) have the most narrow polar pattern in the
front so it may not be the best choice. Some multi-pattern mics have a soft-cardioid (or
sometimes called sub-cardioid) pattern choice which could help with this situation better than the
typical cardioid pattern.

8. Towards the production end of recording vocals, it’s a good idea to have a copy of the lyrics as
your “score” to follow along the performance and take various notes about the musical
performance, technical concerns, etc……also to mark down time-code addresses for the various
sections of the song for easier/quicker access from tape. The notes can help you keep track of
the bad (and good) areas of the performance by having some form of shorthand code to quickly
notate your thoughts for each “take”.
I find it best to wear headphones (even the same type as the singer) while tracking vocals so that
I con concentrate more carefully to every detail of the performance. Headphones have a way (for
me) of excluding other sounds and potentially distracting events. It allows me to keep the sound
and performance focussed inside my head. Wearing headphones also allows you to be in a similar
“space” as the singer which helps in the communication between takes, and the general empathy
between each other during such an artistically sensitive act of recording vocals.

9. Having the vocalist perform in the control room side of the studio can be a good idea as it eases
communication between the engineer/producer and singer. It provides more intimate and

4
© John Klepko 2013

immediate communication than usually occurs when the vocalist is isolated in another room. This
forces you to accept some leakage of the backing tracks from the control room loudspeakers
coming into the vocal mic (or use a tight polar pattern mic, like a dynamic mic might provide)…or
both use headphones.

10. The importance of a really great headphone mix for the singer cannot be stressed enough. Take
the time to make sure that the singer is as comfortable as possible in this essentially unnatural
process of overdubbing a vocal part. They are not usually accustomed to hearing themselves or
the backing music in headphones. Too loud a mix (of their own vocal) will cause them to sing flat,
too soft a mix will cause them to sing sharp – all in an effort to compensate for an inappropriate
mix. Provided the bass track is in tune, it is a good pitch reference to accentuate in the mix,
other elements like piano and guitar although necessary, can add to confusing the pitch center.

11. If you or the vocalist really want to avoid using headphones, then you will have to find a way to
minimize the leakage of sound from monitor loudspeakers. There are 2 methods that can be
used…both are a bit of a hassle to set up and are not very practical, but I present them to you for
further thought, or maybe you can find alternate ways of using these basic ideas. Both of them
involve phase-reversal type signal cancellation to remove the loudspeaker signal at the single
(mono) point of pickup – the microphone.

i. Set up 2 loudspeakers carefully so that they are perfectly equidistant from the microphone
and then, reverse the polarity of one of them. You would send a mono (identical) mix to both
loudspeakers. The combined signals would cancel (for the most part) at the microphone.

ii. Set up a cue mix send to one or more loudspeakers for the vocalist. The placement of the
loudspeakers is not crucial as in (#1). Make sure you are satisfied with the sound, level and
balance of the cue mix, then record (through the vocal mic) a complete “dummy” pass of the song
with the cue mix playing through the monitor loudspeakers. Then proceed to record various takes,
punch-in/outs etc. of the vocal. During mixdown, bring up the initial “dummy” track in the mix, but
reverse the polarity of that channel, if the level is the same, it will cancel out what is common to
the dummy track, and the vocal track…i.e. the loudspeaker mix, and what is left will be virtually
only the vocal.

12. “Comping” is the process of making a “composite” track out of many vocal takes. This process
relieves the singer of trying to get a great performance all in one take; or the engineer of having
to punch-in/out (although this is a totally necessary skill that ought to be developed anyway).
Knowing that comping is possible (provided you have enough open tracks) you can structure the
vocal recording session with the idea of having complete (or near complete) takes of a song. This
would help promote a sense of flow, consistency and story-telling in the vocal of an essentially
aural-only medium. After the singer has gone, the engineer/producer can then assemble the best
words and phrases from the various takes onto one track. This is easy since all the performances
are essentially in sync. Using a ProTools-like recording system for vocals makes this an almost-
enjoyable process. You can effortlessly work from phrase to phrase and drop-in/out from
different tracks or "playlists". Editing gives you the micro-level control of cutting in/out of a
word or syllable with ease. This is not as easy with conventional punch-in/outs. You can, or may
need to EQ and/or compress the different drop-in phrases so that the flow of the line is
maintained. (The different takes may have been recorded at different parts of the day, or
another day for that matter. This would result in sonic differences that need to be ironed out).