Audio Question and
Answer
Q: What does sampling rate refer
to?
A: Whereas analog audio is a continuous stream of audio,
digital audio involves taking 'samples' of audio periodically
and transforming them into binary words. The sampling rate
is the number of samples that are taken in one second. A
sample rate of 44.1 kHz involves one sample every 1/44,100th
of a second.
Q: Why is delay used as an effect?
A: When delays of 35 ms to 40 ms and greater are used, the
listener perceives them as individual delays. When mixed
with the original signal, this can add depth and richness.
If used too much, or on the master track, this can sound
muddy and unintelligable. If shorter delay times are used
(between 15 ms and 35 ms) the delays are too close together
to be perceived seperately. The brain is fooled into thinking
more than one audio track is playing. In effect, this doubles
the overall sound density, without taking up another audio
track. It can be used on background parts to flesh them
out, or on foreground parts to add fullness.
Q: What is the difference between
delay and reverb?
A: Delay and reverb are both used to add richness to a recording.
But while delay adds warmth, reverb adds space to a recording.
Reverb can help to make instruments and vocals sound closer
or farther away from the listener. If used properly this
can make a recording sound live, if done badly it can sound
like total mud.
Q: What is a good bass compression
setting to use?
A: A good starting point is an input/output ratio of 4:1,
with a fast attack (8-20 ms) and a slower release time (1/4
to 1/2 second). You can tweak it from there using your ears
to get the desired sound.
Q: What is popping and how can I
avoid it?
A: Popping is when the microphone picks up puffs of air.
Usually refers to closely miked vocals. Hard consonants
are usually the culprits (p, t, k, ch). Although, closely
miked trumpets can also cause popping. One solution is to
simply move the mic away from the instrument, or vocalist.
Windscreens and pop-filters also help to minimize popping.
Use of an omni-directional mic can also help the problem,
as omni-directional mics are less sensitive to popping.
Q: What is sibilance and how can
I avoid it?
A: Sibilance is an overpronounced pick up of airy consonants
(s, sh, f) in vocals. You can either move the mic further
away from the source (or up or down from the source), or
use a de-esser effect. De-essers are commonly found in stand-alone
processors or as plug-in effects in software recording programs.
Also , you could rotate the head of the mic away from the
vocalist at an angle and/or use a pop filter or windsceen.
Q: What is proximity effect and how
can I adjust for it?
A: Proximity effect is the tendency of directional mics
to boost the pickup of certain bass frequencies as the signal
source gets closer to the mic. The closer to the mic, the
more pronounced the effect. Proximity effect can be greater
for a figure-eight(bidirectional) pattern mic than a cardioid
mic. Most figure-eight mics have a switchable low frequency
rolloff filter, to restore a more natural sounding pickup.
Sometimes vocalists actually prefer the sound of a mic close
up with proximity effect. A more acurate pickup for vocals
is not always desirable. Certain mics are known for their
"sound" for vocals. The MD431 is used frequently
in live broadcast and concert situations for it's "sound".
It exhibits a very desirable proximity effect for a lot
of male vocalists, and also gets rid of a lot of sibilance
and popping of consonants. Because of this, it's sometimes
used for recording situations.
Q: What is "masking"?
A: Masking is a result of the interference and fusion of
two seperate sounds in a similar frequency or harmonic range.
The louder of the two sounds "masks" the quieter,
but only if they contain enough similar frequencies or harmonics.
Two different sounds sources, both between 3600 Hz and 4000
Hz, will tend to blend together. The louder of the two sounds
will drown out the quieter. However, a louder sound at 3000
Hz will have to be a lot louder in comparison to drown out
a 1000 Hz sound. Masking is one of the many reasons mixing
and equalization are so important in the recording process.
When mixing and stereo panning don't get rid of masking
problems, usually proper eq fixes the problem. If you boost
an eq frequency on one sound(track) and cut the same frequency
on the other offending sound(track), the masking is usually
a lot less noticable. Masking can actually be somewhat desirable,
such as in different orchestral sections. If you want a
whole section of violins to blend together, masking can
help blend the section together as a whole. But when flutes
and violins sound like they're muddling together, you have
a masking problem.
Q: What is a frequency response curve?
A: A frequency response curve is the graph of a microphone's
response across different frequencies. The frequency is
graphed in the x-axis, and the amplitude(volume) is graphed
on the y-axis. If you have a "flat frequency"
response, this means there are no weak frequencies or boosted
frequencies in the overall pickup of the microphone. If
you graph a flat frequency response, it will look like a
horizontal line across the whole frequency range. A mic
with a flat frequency response will be very acurate. Good
for miking some things, not good for miking others. If you
don't want to pick up every nuance of the tone(such as with
an imperfect vocalist), you probably don't want a mic with
a perfectly flat frequency response. Sometimes microphones
have frequency response curves engineered into them to get
a certain desirable "sound". Most dynamic mics
don't have a flat frequency response. Most mics with an
almost perfectly flat frequency response are small condenser
microphones.
