Digital Signal Processing
Digital delay is really where the more interesting digital signal processing effects happen among which are echo, reverberation, chorus, and vibrato to name a few. Digital delay is accomplished by using a temporary or short term digital memory where the digital information represented by the audio signal gets stored for a while. Retrieving the contents of the memory after a time lapse results in a delay.
 |
Figure 3, above, simplified block diagram of a digital delay suitable for echo production.
In figure three above, you see a block diagram of a digital signal
processor set up for tape echo simulation. The signal enters at the left
and goes through the Analog to Digital converter to emerge as digital
data which is then stored in a Random Access Memory. After a short while,
the memory contents are read out and pass through the Digital to Analog
converter from which the signal emerges to go on to subsequent amplifiers,
speakers and/or recording equipment. The whole is under the control of
the central processing unit which establishes the sample frequency and
other parameters necessary for the correct operation of the system. Note
that there is also a feedback with adjustable gain where the digital information
at the output of the RAM can be feed back to its input but at a reduced
level.
Among the many advantages that this has
over tape echo is that everything is accomplished without any moving parts.
The only moving parts are just the control knobs or buttons by which the
musician sets the levels, delay time and feedback as necessary to get
the effect he wants for a particular instrument or piece of music. This
of course makes it essentially maintenance-free in that there is no continuous
tape cartridge or tape reel to worry about.
A number of different delay lines can be
combined having different delay times and feedback rates which can then
simulate reverberation. Because this all happens on small chips instead
of in large electro-mechanical devices, it's possible to make very complex
echo and reverb patterns that can simulate the reverb or acoustical characteristics
of many different types of rooms, so that the apparent acoustical environment
can be made to sound like anything from a bathroom to a large concert
hall. Digital reverb processors can also simulate the earlier spring type
reverbs that were in use prior to the advent of these units. They can
also create reverse reverb, where the reverb starts out quietly and gradually
gets louder, as well as the so-called gated reverb where the reverb goes
on at a constant level for a second or so and then cuts out.
One of the many useful features of digital
signal processing is its ability to change the pitch of all signals which
appear at the input of the A/D converter. On the following pages are conceptual
diagrams or animations which show, in a graphical and analog format, how
it is possible to shift the pitch of a signal. Pitch changing is a delay-based
function. Digital pitch changing can be very subtle, where it creates
a nice "live" ambient effect. This is particularly helpful on
some elec-tronic organs where their absolute tuning accuracy and complete
lack of any small and subtle pitch discrepancies between the tempered
scale intervals and true harmonics can make the tone sound somewhat dull
and lifeless. Replicas of the input signal can be shifted very slightly
in pitch both above and below the actual pitch and then combined in stereo
channels.
By shifting the pitch continuously above
and then below the input pitch, we can create a pitch vibrato effect,
and because digital signal processors are all solid state, chip-based
devices, we can easily produce very complex vibratos by having numerous
different, out of phase vibratos acting simultaneously. By doing this,
and by creating a stereo signal from a mono signal, and by also adding
a slight stereo tremolo to the result, we can produce very credible imitations
of the effect you get with a rotating Leslie speaker.
