Reverberation
Over the course of the last couple of lectures, we've looked at audio cues with regard to distance
(essentially one-dimensional spatial placement), and left-right positioning (essentially twodimensional spatial placement). In today's lecture, we're going to look at reverberation (or reverb
for short).
What is Reverberation?
Reverb is a natural acoustic effect, which we hear particularly well in large enclosed spaces.
Essentially, reverberation consists of thousands of discrete echoes which arrive at our ears
having first reflected off the various surfaces in the space, and fuse together to create an acoustic
halo of sound.
It is important to note, however, that the sound that reaches our ears is a mix of the direct (dry)
sound, and the reverberated (wet) sound (see Figure 1). The exact balance of the sound will
depend on the distance of the source from the listener, the construction of the room, and the
intensity and duration of the sound. If a sound to which artificial reverberation has been applied
is heard without any of the original signal present, it will sound muddy and un-natural.
Figure 1 - Schematic representation of reverb showing both dry and reverberant elements.
It's very tempting to say that reverb is a series of echoes, but this isn't quite correct. 'Echo'
generally implies a distinct, delayed version of a sound, as you would hear with a delay more
than one or two-tenths of a second. With reverb, each delayed sound wave arrives in such a short
period of time that we do not perceive each reflection as a copy of the original sound. Even
though we can't discern every reflection, we still hear the effect that the entire series of
reflections has.
For a typical reverb, for a short period after the direct sound, there is generally a set of welldefined and directional reflections that are directly related to the shape and size of the room, as
well as the position of the source and listener in the room. These are referred to as the early
reflections.
After the early reflections, the rate of the arriving reflections increases greatly. These reflections
are more random and difficult to relate to the physical characteristics of the room. This is called
the diffuse reverberation (or sometimes the late reflections).
The diffuse reverberation is the primary factor establishing a room's 'size', and it decays
exponentially in good concert halls. In figure 2 below, we present a typical impulse response for
a room:
Figure 2: Impulse response of a room.
Characterising reverbs
One of the most important characteristics of a reverb is the reverberation time. Strictly speaking,
the reverb time is the amount of time it takes for sound pressure level or intensity to decay to
1/1,000,000th (-60 dB) of it's original value. Longer reverberation times mean that the sound
energy stays in the room longer before being absorbed, which generally indicates a sense of
space and perhaps a room that is constructed from relatively reflective material.
The surfaces of the room determine how much energy is lost in each reflection. Highly reflective
materials, such as a concrete, tile, brick walls, and windows, will increase the reverb time as they
are not particularly absorbent. Absorptive materials, such as curtains, heavy carpet, and people,
reduce the reverberation time. You can check this using a lecture theatre - turn up early before
the room starts to fill up with bodies, and the room will sound 'bigger', but once the lecture is
underway, the sonic size of the room will seem to diminish. Why? It's because people tend to
absorb quite a bit of energy, which in turn, reduces the reverberation time.
Size also has a great effect on the reverberation time - bigger rooms tend to have longer
reverberation times since, on the average, the sound waves travel a longer distance between
reflections (which is where most of the energy is lost). The air in the room itself will also
dampen the sound waves, reducing the reverberation time, and as we mentioned a couple of
lectures back, this atmospheric absorption varies with air pressure, humidity and temperature,
and high frequencies are affected most.
Since we are so accustomed to hearing reverberation, we often have to specifically listen for it in
order to notice it, and in fact, such is the commonality of the effect in nature that it is really only
noticeable when we don't hear it. If any of you ever have the opportunity to visit an anechoic
chamber (at a really high-class recording studio, say), then please do so - they are quite unusual
spaces.
Another important parameter with regard to reverberators is the predelay. This is the amount of
time before the first reverberations of a signal are heard, i.e. the time before the first early
reflection in the impulse response. In some cases, the predelay may be defined as the time before
the late reflections are heard. More complex reverberation units may actually allow you to set the
predelay for both the early and late reflections. For simulation of real environments, the predelay
for the early reflections should always be smaller than for the late reflections.
Recreating reverb artificially
The earliest attempts at artificial reverberation of recordings took place in specially constructed
acoustic echo chambers. These were separate rooms with a loudspeaker at one end and a very
high-quality microphone at the other. The sound to be reverberated was played over the
loudspeaker, picked up by the microphone, and mixed back with the original signal.
Echo chambers offer a unique acoustical ambience created by one specific room, loudspeaker
and microphone. When these work together, the quality of reverberation is excellent. Indeed, this
technique is still employed by some studios today. The main drawback to this approach is, of
course, that it is nigh on impossible to alter the reverb characteristics.
The more usual approach when adding reverb is with a totally artificial reverberator. The first
attempts to synthesise reverberative effects were huge electromechanical devices containing a
reverberating medium like a spring or a metal plate - in fact, cheap guitar amplifiers offering
reverb still use this technique. Try kicking one over and listen for the 'twaaaang!!!'
They were also used in early Hammond organs and have a very artificial, but quite characteristic
sound.
The operation of a spring reverb is pretty simple - the audio signal is coupled to one end of the
spring by a transducer. This creates waves that travel through the spring. At the other end of the
spring, there is another transducer that converts some of the motion in the string into an electrical
signal, which is then added to the dry sound. When a wave arrives at an end of the spring, part of
the wave's energy is reflected and stays in the spring. It is these reflections that create the reverb
characteristic sound.
Often you will find several springs being used together in a reverb unit. Each spring can be of a
different length or under a different tension to avoid the uniform behaviour in a single spring,
where all the reflections occur at fixed times. In a sense, it increases the 'randomness' of the
echoes. However in most reverb units, the spring lengths and tensions are fixed in the design
process, and not left to the user to control.
Plate reverbs are not used widely outside of the studio since they are expensive and rather bulky.
The setup is similar to a spring reverb, but instead of being connected to the springs, the two (or
more) transducers are connected to different points on a metal plate. These transducers send
vibration waves throughout the plate, and reflections occur each time a wave reaches the edge of
the plate. The reverberation can be controlled by adjusting the damping of the plate and the
location of the transducers.
Digital reverb units vary greatly in complexity and in quality. In addition, the editing options
vary greatly according to the pedigree of the unit. However, all but the most basic reverb units
offer the ability to alter the reverberation time, the length of time taken for the reverberated
signal to decay by 60dB from its peak amplitude; the reverb pre-delay, the length of time
between the direct sound and the first early reflections reaching the listener; the room size, and
the reverb type, for example concert hall, tiled room, garage, canyon, etc. Digital reverb units
range in price from a couple of hundred pounds to the very best Lexicon units which retail in the
hundreds of thousands.
Digital reverberators use time delays, filters and mixing to achieve the illusion of sound
scattering within a room. From a signal-processing standpoint, a reverberator is a filter with an
impulse response that resembles that of a room.
Effective simulation of reverb requires a high echo density. Some early digital reverbs produced
as few as 30 per second, whilst in actual spaces, densities of more than 1000 echoes per second
are typical. Many of today’s reverberators provide a control that allows you to adjust the echo
density to suit a desired effect, from discrete echoes right through to a dense, fused reverb
pattern.
The discrete early reflections of a concert hall can be simulated by means of a tapped delay line –
a delay unit that can be ‘tapped’ at various points to put out several versions of the input signal,
each of which is delayed by a different amount.
Again, I don't want to get into the algorithms employed, but as a general rule of thumb, a digital
reverberator works by using delays and filters to produce a series of echoes, each of which
cascades to produce more echoes. By repeating the process several times it is possible to create
dense echo patterns that simulate the effects of natural reverberation.
Why use Reverb?
If reverb is always around us, why do we add reverb to recorded sounds?
Well, the principal reason is, of course, flexibility. By recording instruments 'dry', we can, at a
later date, apply artificial reverberation to simulate the placement of such sounds in artificial or
virtual spaces.
Also, with the increase in power and complexity of sound synthesis, it is possible to create entire
soundscapes synthetically. Such sounds have no natural reverb. Even some types of musical
instrument, such as the electric or bass guitar, will have no natural reverb.
A very dry signal can sound quite unnatural, particularly when mixed with sounds that have a
natural reverberant characteristic.
So how should I approach this whole reverb thing?
As always, learn by experience. I'd suggest booting up SoundForge, which has a reasonably good
reverb processing unit built into the package itself. Load up a typical dry sound (impulse-type
sounds tend to give the most apparent results - try a snare hit or a bass drum thud) and fiddle.
Try each of the different reverb algorithms that are on offer (there are loads!) and see how each
one subtly changes the character of the sound. Alter the reverb time and the wet/dry/early mix –
what effect does this have on the perceived space?
Finally, think hard about the sorts of thing we've been talking about over the last few days. Can
you bring them all together in a meaningful way? Can you apply dynamic filtering effects, for
example, and in what situations would this be useful? How does reverberation help us to locate
sounds spatially? Can you put this to work with the stuff we've already done regarding spatial
positioning and distance location?
Give it some thought.
© Kenny McAlpine 2002