 |
|
|
Okay, you’ve got some vocal
tracks down. Through the extensive use of candles,
mood lighting, and some encouragement, you’ve
managed to record (and maybe sing) a vocal take
that captures just the right emotion. That’s
the hard part. So what now? What can be done to
make the most of your voice recordings?
Well, a lot really, and actually
now more than ever. Digital audio has really allowed
people to manipulate voice recordings in ways they
could not dream of before. And that I believe, in
many ways has allowed singers to worry more about
capturing the right feel of a take rather than just
hoping for one that’s in tune all the way through.
But what exactly can be done and how? I cover what
I believe to be the three major ways that one normally
manipulates voice, fixing, enhancing, and adding.
Fixing — or "Making
the best of a good situation"
I would honestly like to know
if any record in the Billboard Top 200 (not counting
live records) has a voice recording which did not
have at least some "correcting" going
on. I doubt it. Pop records (and by this I mean
all popular records in all genres) rely on vocals
for a lot of their appeal. And people of my generation
or younger have been raised on records with "flawless" vocal
recordings. So what are the ways that people correct
vocals? Let’s look at a few, some high-tech,
some not so.
"Comping"
This is by FAR the most common
technique employed for enhancing performance and
has been in use long before digital audio, however
Digital Audio Workstations were made for this type
of work. Comping is basically the technique of recording
several takes of a vocal performance and taking
the best parts to create one unique performance.
This differs from just punching in to fix parts
in that the vocalist will perform the entire parts
all the way through. Often it goes beyond just making
sure that one take is in pitch, but goes to which
take has the best emotional turn of phrase. I spoke
to an engineer who had worked on all the Chris Isaac
records and he said it was not unusual for Chris
to do 100 takes for each song.
"Popping ‘P’s"
Also often called "plosives",
these are caused when certain consonants cause air
to flow over the mike in such a way that creates
noise. These can happen even when care is taken
to employ limiters and pop filters. Sometimes they
can be fixed by careful editing of the waveform.
Plosives often produce a clipped waveform that can
be redrawn in without affecting the sound tremendously.
You can also sometimes edit the errant part out
if you are careful to take a small enough piece
and you replace it with silence or a non popped ‘P’.
Often the best solution is to just retake the vocal
part.
Sibilance
This is that hissing sound
that can be produced especially when the material
contains a lot of ‘s’s. This is usually
fixed by employing a de-esser. (I bet you’re
thinking, "wow, what a surprise"). A de-esser
is usually a dedicated frequency specific compressor
that focuses in on that part of the frequency spectrum
where esses hide. However, any multiband compressor
can be used to the same effect. Basically, whenever
anything in that frequency range gets louder than
the rest of the material it is compressed, meaning
just the sound in that frequency is turned down,
just for that moment. It accomplishes the same thing
as if you used an EQ and turned down that range
by just the right amount at just the right time,
if your hands and your ears were fast enough.
Although Deessers are usually
fairly simple to use, care must be taken to use
only the amount that is necessary to fix the track
lest you remove high end "sparkle" from
the track making it sound dull. It is also a good
practice to only employ it on the section of the
performance where it is needed.
An example of a hardware de-esser
is the Focusrite
VoiceMaster and an example of a software de-sser
is the Renaissance DeEsser in the Renaissance
Collection 2.
Pitch Correction
Here is where we have seen
big advances over the last couple of years. New
software has allowed engineers and producers to
fix small (or even large) intonation problems without
affecting the rest of the performance. This can
be done automatically by defining a preset scale
to correct to, or manually to going in and fixing
particular notes. This manual method has also unleashed
the now famous (or infamous) Cher "Believe" effect
where pitch correction software was intentionally
overused. However, normally pitch correction can
be done so it is unnoticeable, but extreme care
must be used to apply the effect properly to avoid
noticeable artifacts.
Here are a few tools that
allow you to do pitch correction:
- Antares
AutoTune — really
the only kids on the block
until recently, Autotune
by Antares revolutionized
the way people thought
about pitch correction.
AutoTune is a plug-in that
is available in most formats,
so you can probably use
it within whatever Digital
Audio application you are
using now. AutoTune comes
in 2 flavors, the LE version
which only allows you to
do automatic pitch correction,
which I have found to be
quite enough in most cases,
or the full version. The
full version allows you
to manipulate individual
notes. Another feature
is that because the process
of removing errors in intonation
often removes vibrato,
you can add back in vibrato
and adjust the width and
timing to make it realistic.
And of course nothing prevents
you from adding vibrato
to tracks that don’t
need pitch correction.
Antares also makes a hardware
version of Autotune for
people who prefer their
software to come in a rack
mount box.
- Celemony
Melodyne — While
Melodyne can do much more
than pitch correction,
it does pitch correction
extraordinarily well. Like
the full version of Autotune,
it allows you to manipulate
individual notes, but unlike
Autotune it treats vibrato
and pitch as separate items,
so you can correct the
pitch of a note without
affecting the vibrato.
Melodyne also accounts
for transitions between
notes that when a singer "swoops" up
to a note (a glissando
for the educated among
you) this is also allowed
for and not "corrected".
For pitch correction, Melodyne
is an outstanding tool,
however, it is not a plug
in, so you must extract
any audio you wish to process,
process it in Melodyne,
and then place it back
into place. Although I
found this relatively easy
to do, there may be many
situations where this is
simply not practical.
- Spectral
Design Voice Machine — While
I am not personally familiar
with this product, it
does purport to have
many of the pitch correction
features of the previous
products, and it is by
far the least expensive.
For an in-depth review
of this just released
product check out Ian
Bruce-Douglas's or Miriam
Kolar's review.
Profanity Removal
Admittedly this next tip may
have limited appeal, but since I use it regularly,
you may find it useful. Hip-hop lyrics often contain
a lot of profanity that must be removed before it
can be played on the radio or sold at Walmart. Sometimes
a "clean" version will be sung/rapped,
sometimes it won’t. When it isn’t, it
is the responsibility of the producer to create
a clean version and here is what I do. I create
a sampler instrument with the sound effects I want
to use to cover up the offending words (this is
much more musical than just "bleeping" out
the words). Then I sidechain a compressor to this
instrument so that when this instrument plays the
vocal track (often on a bus) is turned all the way
down. This technique is called "ducking" and
is often used on kick drum/bass to increase the
amount of "tightness". Then I can "play" the
sound effects so that they fit in with the rhythmic
flow of the song. And I don’t need to save
another version of the song, when I want to have
the original "unclean" version, I just
mute the sound effects track.
Enhancing Voice — or "you
can’t have too much of a good thing"
So now your track is perfect,
in tune, just the right notes held for just the
right amount of time. What can you do to "flatter" that
vocal take and to make the most of it? Here are
a few tips:
Doubling
Ok, this is the ultimate low-tech
tip, which a lot of people already know, but it
is also one of the most effective. The same performance
is sung twice and both performances are played,
often with one lower in volume than the other. Often
it is more effective if sung in a different "voice" (e.g.
softer, louder, with more breath, etc). Of course,
if you had to spend a lot of time correcting the
vocal take this may not be a simple as it sounds,
but doubling produces a very nice, and very natural "thickening".
Now some of you with digital workstations may think, "I
already have one perfect take, all I have to do
it cut and paste and bingo, there’s my double".
Good thinking, but unfortunately, incorrect. Takes
that are exactly the same will not produce a nice
doubling effect but will just give you a chorusing
effect, which can probably be better accomplished
with a chorus or delay. Many popular vocalists "sound" is
defined by how they layer their voice in the studio.
Reverb
Again, not exactly a news
flash for a lot of people, but such a big and important
concept that it is too much to cover here. Let me
just say that for vocal-centric music like Country
and R&B, marrying the correct reverb to your
vocal take is one of the most important decisions
you can make. Just remember that about 80% of what
you hear in the real world is reflected sound, so
don’t think of Reverb as an "effect",
but as part of the instrument. One thing to consider
as well is the "emotional proximity" of
the voice. Vocals that are meant to be intimate
should be relatively dry to produce a "close
up" effect, so it sounds like the singer is
right in the room with you. Vocals which are "bigger",
should have a larger room applied to them so the
illusion is more of seeing a performance in a large
concert hall.
Chorus/Delay
Proper amount of chorus or
delay can produce some nice and natural effects.
I find that very slow and very light chorus or delays
between 20 and 40ms can produce a nice effect without
sounding effected.
Compression
Compression is usually critical
to a vocal mix to make sure that the vocals "sit" in
the mix, meaning they don’t sound separate
from the rest of the music. Remember also
that compression will enhance the highs and lows
of anything being compressed, especially the lows,
so if want to bring out the low end in a voice,
compression can be a handy tool. Also, many analog
compressors can add "warmth" to a track
by increasing the quantity and the quality of the
midrange. Vintage compressors like the LA2A or 1176
are renowned for their ability to add warmth to
tracks, and much of the same effect can be produced
by their software cousins. Take a look at their
much easier to find and significantly cheaper software
reproductions produced by Bomb Factory or Universal
Audio on their powered Plug-ins card.
Adding — "dialing
in the shoop in the shoop
de woop de woop"
Now that your lead vocal is
rich, thick, warm and perfect, it’s time for
a little company. The history of Pop music would
be a much colder and darker place were it not for
the "ooooh"’s, "ahhhhh"’s,
and "That’s right sister"’s
of the backup singers. And in the current round
of R&B and girl/boy groups, the line between
lead and backup vocals has become blurred, much
of the lyrics are sung in 3 or 4 part harmony. So
what do you do if you’re just Beyonce, with
no Kelly and Michelle around? (or Diana with no
Mary or Florence for you older cats)
In steps technology. Here are
a few tools for creating rich background textures:
- Celemony
Melodyne— By
far the easiest and
most powerful of all
the tools is Melodyne.
My complete review
of Melodyne itself
is available
here. Melodyne
is to vocals as the
word processor is to
words. You can create
background vocals in
any amount and in any
configuration up to
however many tracks
of audio your system
supports. You
merely cut and paste
your original vocal
track and edit these
copies to be whatever
you want, in time,
pitch, vibrato or formant
(see Ian
Bruce-Douglas’ review
of Voice Machine for
an excellent explanation
of formant). Whatever
you can imagine you
can do. The upside:
complete flexibility,
the downside: no presets.
You need to musically
be able to work out
the harmonies; Melodyne
does not give you a "head
start".
- Spectral
Design Voice Machine — Designed
more specially for
just creating harmonies
is Voice Machine. It
allows you to use one
voice as the key and
create harmonies by
playing midi notes
on a keyboard. This
method is unarguably
faster than working
with Melodyne that
requires you to export
your audio. The price
of Voice Machine also
may make it more attractive
to people who may not
require all the power
that Melodyne provides.
- TC
Helicon Voice Prism — The
Voice Prism from TC
Helicon is hardware
box that provides features
similar to Voice Machine,
plus a Mic Pre, a compressor,
EQ, Gate, and Effects.
By far the most complete
as it obviously provides
features that no software
can provide, it also
has extensive capability
for creating harmonies,
either using presets,
or by creating melodies
using a MIDI keyboard.
We did not receive
our test unit in time
for a hands-on review
but based upon the
specs and the review
in Pro Audio Review,
this is an impressive
unit that is both easy
to use and powerful.
Check back soon for
our complete review.
If you have any questions or
comments about Voice Manipulation, please email
me at brent@audioMIDI.com
|
|
