Until rather recently in television far more attention was paid to video than to audio. "Good sound" was when you could make out what was being said; "bad sound" was when you couldn't.
This has changed. With the advent of stereo, 5.1 surround-sound, and home theaters, audiences have much greater expectations.
Before we can discuss some of the basic audio production concepts, sound, itself, must be understood.
Sound has two basic characteristics that must be controlled: loudness and frequency.
Although sound loudness is commonly measured in decibels (dBs), that term actually refers to two different things.
First is dBSPL (for sound pressure loudness), which is a measure of acoustic power. These are sounds we can directly hear with our ears.
These decibels go to and beyond 135, which is considered the threshold of pain and, by the way, the point at which permanent ear damage can occur.
If your ears "ring" after being around a loud sound, this should be a warning sign that you have crossed the threshold of potential hearing damage. (The damage, which is irreversible, often goes unnoticed, which probably explains why the average 50-year-old in some countries has better hearing than many young people in the U.S.)
Musicians who must be around high-level sound use musician's plugs -- special earplugs that attenuate sound level without distorting the frequency range. In case you are thinking about starting your own rock band, HEAR, (Hearing Education and Awareness for Rockers) at hearnet.com has more information.
Various sound pressure decibel levels (in dBSPL's) are shown here.
The second use of the term decibel, dBm (for the milliwatt reference level) is a unit of electrical power.
In audio production we are primarily interested in dBm, which represents levels of electrical power going through various pieces of audio equipment.
Two types of VU meters for measuring the loudness of sound are in wide use: the digital type and the analog type.
The 0 to 100 scale on the left side of this illustration indicates modulation percentage (percentage of a maximum signal), and the scale on the right is in dB's.
Contrary to what logic might dictate, 0dBm (generally just designated 0dB on a VU meter) is not "zero sound" but, in a sense, the opposite, the maximum desirable sound level. (Granted, that's a bit confusing, but, then again, we didn't make up this system!)
The 0dB point on the meter is just a reference point. Therefore, it's possible to have a sound level on the meter that registers in negative dBs, just as it's possible to have a temperature of -10 degrees Centigrade or Fahrenheit.
The animated versions above illustrate how digital meters respond to sounds.
The VU meter on the right is the traditional analog meter that has been around in one form or another since the dawn of radio.
Although easy to read, most versions do not accurately respond to short bursts of loud sound.
The dB level going through audio equipment must be carefully controlled. If the signal is allowed to pass through equipment at too low a level, noise can be introduced when the level is later increased to a normal amplitude (audio level).
If the level is too high (significantly above 0 dB or into the red areas on the VU meter), distortion will result -- especially with digital audio. To ensure audio quality, you must pay constant attention to maintaining proper audio levels.
The animated meter shown here indicates a sound
level that is a bit too high. Ideally, the needle should not go deeply
into the red area this often.
Frequency relates to the basic pitch of a sound -- how high or low it is. A frequency of 20 Hz would sound like an extremely low-pitched note on a pipe organ -- almost a rumble.
At the other end of the scale, 20,000 Hz would be the highest pitched sound that most people can hear, even higher than the highest note on a violin or piccolo.
Frequency is measured in Hertz (Hz) or cycles per second (CPS). A person with exceptionally good hearing will be able to hear sounds from 20-20,000 Hz.
Since both ends of the 20-20,000Hz range represent
rather extreme limits, the more common range used for television production
is from 50 to 15,000 Hz. Although it doesn't quite cover the full range that
can be heard by people with good hearing, this range does cover almost
all naturally occurring sounds.
The Frequency-Loudness Relationship
Even though sounds of different frequencies may technically be equal in loudness (register the same on a VU meter), human hearing does not perceive them as being of equal strength.
The red line on the graph (roughly) shows the frequency response of the ▲human ear to different frequencies.
Because of the reduced sensitivity of the ear to both high and low frequencies, these sounds must be louder to be perceived as being equal to other frequencies. (A much more detailed version of the relationship between and audio frequency and perceived loudness is available here.)
You'll note that a good-quality microphone (the
green line) is relatively "flat" in the all-important 50-15,000 Hz.
Equipment and listening conditions also greatly affect how different frequencies will be perceived. To compensate for some of these problems, we can adjust bass and treble controls of playback equipment.
More sophisticated equipment will include a graphic equalizer, which goes a step further and allows specific bands of frequencies to be individually adjusted for loudness.
A graphic equalizer may be necessary to help match audio segments recorded under different conditions, or simply to customize audio playback to the acoustics of a specific listening area.
Note that the graphic equalizer shown here can control nine specific frequency areas (bands).
Any piece of audio equipment -- microphone, amplifier, recorder, or audio speaker -- can adversely affect the fidelity of sound. However, it's the microphone (the initial device that transduces sound waves into electrical energy) and the audio speaker (the device that changes electrical energy back into sound waves) that represent the weakest links in audio quality.
To some degree it's possible to use graphic equalizers
and similar audio equipment to "clean up" the frequency response
of a poor microphone. However, even the most sophisticated audio techniques
can't work miracles. Thus, the better the original audio signal, the better
the final product will be.
Sound, both as it's recorded and played back, is more affected by the acoustics of a room or studio than most people realize.
In an effort to create totally soundproof studios, early radio stations used to use thick carpets on the floors and heavy soundproofing on the walls.
Although possibly successful as soundproofing, the result was a lifeless and dead effect that we're not used to hearing in a normal listening situation, such as our living room. Therefore, a slight bit of reverberation is both desirable and realistic.
Two types of soundproofing material are shown on the left.
A room with a tile floor and hard, parallel walls will reflect sound so much that it interferes with the intelligibility of speech. Sometimes it's desirable in these situations to place free-standing sound absorbing items in the room -- things like sofas and rugs -- to break up sound reflections and reduce reverberation.
The ideal room for recording or listening to sound has just enough reverberation to sound realistic, similar to your living room, possibly, but not enough to reduce the intelligibility of speech.
Issues Forum Author's Blog/E-Mail Associated Readings Bibliography
Index for Modules To Home Page Tell a Friend Tests/Crosswords/Matching