|
Although analog recording still
plays a major role in audio production, digital
audio systems are commonly used for recording, editing,
mixing, adding effects, mastering and scoring. Analog
recording suffers from limitations such as saturation,
tape noise, flutter, generational loss and expensive
hardware. Where as digital audio offers pristine
audio quality, powerful digital signal processing
(DSP), near-perfect duplication and affordable hardware/software
solutions. For these reasons many commercial and
home musician studios have chosen digital audio.
Analog recording uses audio
tape, which is a very thin strip of plastic, coated
with magnetic particles. The recording process aligns
the magnetic particles into a pattern on the tape
that represents the electrical signal. This creates
a continuous representation of the electrical signal.
When acoustic or analog audio
is represented as a wave, it typically looks something
like this:
If we were to lay a grid over
the sound, we would see that acoustic waves are "smooth",
meaning they have an infinite number of points on
the grid.
The situation is slightly different
when it comes to digital audio.
Computers have a hard time dealing
with things that are infinite, so when recording
an analog signal into a digital system, the electrical
signal must be converted into a language that computers
can understand. This language is called Binary.
Binary data (bits) consists of just 1's and 0's
in series and the translation of audio into this
numerical language is accomplished through the use
of an analog-to-digital converter (A/D). Some computers
have basic A/D converters built-in, but for high-quality
recording additional audio hardware is available.
When digital audio is played from the digital system,
the binary data is converted into an electrical
signal (so your audio speakers can monitor the signal)
by a digital-to-analog converter (D/A).
Sampling
The process that an A/D converter
uses to capture an analog signal is called sampling.
Sampling is like taking a snapshot of the sound
wave. This creates a stepped representation of the
electrical signal. When computers digitize sound,
the computer takes snapshots of the sound at fixed
points. A digitized version of the above wave might
look something like this:
While the shape is similar to
the analog wave, they are not exactly the same.
The more often that the computer takes those pictures
of the sound, the closer to the original it will
sound. The number of times per second that a computer
takes that snapshot is refereed to as the "Sampling
Rate ". The higher the sampling rate, the more samples
taken per second. There are many different common
sampling rates that you might see:
- 44.1
kHz = 44,100 Samples per
second - This is the standard
for CD's
- 48 kHz
- 88.2
kHz
- 96 kHz
Bit Depth
The other aspect of how computers
digitize the audio is the "resolution" or "bit rate"..The
resolution of a sample represents its dynamic range,
or levels of volume. The higher the resolution,
the greater the dynamic range; and the lower the
resolution, the less the dynamic range. For example,
a 1-bit system can only represent two levels of
volume - off or on (0 or 1). A 2-bit system can
represent four levels of volume - off, soft, medium
or loud (00, 01, 10, 11). A typical resolution is
16-bit, which is the resolution used for Audio CDs.
A 16-bit resolution can represent 65,536 (or 96
dB) levels of dynamic range. A 32-bit system can
represent 1,048,576 (or 192 dB) levels of dynamic
range. That is over one million levels of dynamic
range!
The different common bit rates
are:
- 16 Bit
- This is the standard for
CD's
- 18 Bit
- 20 Bit
- 24 Bit
Digital Recording capabilities
are often listed as Bit Depth / Sample Rate.
So the standard recording for
CD's can be represented as being 16 Bit/44.1 kHz.
Most high end audio software now can record at 24
Bit/96 kHz.
| 
