The Parameters of Sound

Data sonification commonly involves taking data—numeric values—and assigning those values onto
sound parameters to highlight particular aspects of the data. Making choices about the ordering and
controlling of sound parameters over time is as old as music notation. In fact, standard music notation
highlights certain aspects of sound in its writing. For example, music notation outlines what notes should
be played, when the notes should be played, how long the notes should be played, how loud the sound
should be played, and if noted, the character of the sound and any changes over time (see Figure 1 as an
example of standard music notation). The instrumental part partially dictates the timbre, or color, of the
sound. John Cage, in his 1937 treatise breaks down sound into four various components: frequency,
amplitude, timbre, and duration (Cage 1961: 4).

In the 1950s, when Max Mathews and colleagues were busy creating computer-assisted music with their
MusicN series at Bell Labs, understanding the various components of sound was important for electronic
sound synthesis (Chadabe 1997). Today, there are vast numbers of digital tools and methods for mapping
data onto sound, but the parameters of sound are quite the same, and the choices of mapping are still
very much designed. It’s important in sonification to remember the fundamental parameters of sound,
even if one is thinking of mapping the sonification in musical terms (say musical pitches and orchestral
instruments), as data narratives (Siu et al. 2022), or even as part of virtual audio environments.

At the University of Oregon, where students and faculty build their own digital musical instruments,
compose, and perform on these instruments as part of a digital musical practice (Stolet 2022), we often
rely on five sound parameters in mapping data streams. I like to state them as Frequency, Amplitude,
Timbre, Location, and Duration, or simply FATLD. As John Cage would later describe in a 1957 lecture,
sound parameters have envelope shapes — a frequency envelope, an amplitude envelope; yet, an
envelope describes the temporal changes of sound, changes of one or more sound parameters.
Certainly, there isn’t anything wrong with adding Envelope as a sound parameter, making the acronym
FATLED. Or maybe FDELTA?

Regardless of the arguments around what should or should not be included as a sound parameter, these
five, if not six, fundamental sound parameters are critical to mapping in data sonification. Sound is the
carrier of information in sonification, the "path of information exchange, not just the auditory reaction to
stimuli” (Truax 2001: xviii). Below, I describe each sound parameter and provide sound examples. The
examples have less to do with communicating a message and instead highlight a given parameter. Each
example is driven by a pseudo-random number generator controlling the given parameter while all other
parameters remain static. The goal of listening is to hear the isolated sound parameter.

Frequency

Frequency is often described as the periodicity or the number of vibrations in a second. When one
plucks a string, the number of times that string vibrates in a second corresponds to its fundamental
frequency, and if we hear that sound, we may even be able to describe the vibrations as a musical pitch.
For example, the musical pitch for concert tuning is A440, or A is equal to 440 cycles per second (cps).
You can think of pitch as perceived frequency, even though a musical note contains many more
frequencies than its fundamental or base pitch. The human ear can detect (at its best) a range between
20 Hz to 20,000 Hz.

Amplitude

Amplitude is an attribute of a soundwave and refers to the "maximum displacement" or height of the
wave (Pierce 1983: 41). If we have two sounds with the same shape (i.e., frequency and timbre), a sound
with a larger amplitude will sound louder and the smaller amplitude will sound quieter. Amplitude refers
to the quality of the sound, while loudness is our perception of the quality of the sound. Our experience
of acoustic sound correlates energy with amplitude. For example, if you strike a drum harder you
transfer more energy into the drum surface, causing the drum to vibrate at a greater amplitude, which
causes larger variations in air pressure. When those pressure waves hit our ears, they transfer more
energy onto our eardrums.

Timbre

Timbre is the quality of sound that makes the sound unique. While frequency and amplitude play a role
in timbre, sounds without a clear pitch, like cymbals, can diverge in timbre. Timbre also consists of
additional factors. As University of Oregon musicology professor Zachary Wallmark describes, there are
"physical, perceptual, and social qualities of timbre" (Wallmark 2022). So while the physical makeup of
an instrument can make it unique to other instruments, a listener can "extract the pattern of a sound"
based upon its timbre, and "timbre is perceptually malleable between people", who are "embedded
within specific historical and social contexts" (ibid.).

Location

Sound comes from somewhere. Location is the position of the sound in a space and in relation to the
listener in space. When we listen to an acoustic bluegrass group (no amplifiers), the musicians are
arranged in physical space, with sounds emanating from their instruments. By changing our position, or
by rearranging the musicians, the location and the balance of the sounds change.

In stereophonic audio (two channels of playback), it is common to describe stereo audio as playback
from left and right. We can pan, or position, a sound between two speakers by using the gain of the left
and the right channels. In this way, the 'pan' control allows one to arrange sounds between the left and
right channels (or as we might perceive the sound, between the left and right ears).

Duration
Duration is the length of a sound. In music, there are specific values that describe a sound's duration, for
example, a quarter note is a sound that is to be played for one beat in length. Musical duration is
dependent upon tempo, but sounds in sonification may not have a tempo, unless sound events are
synchronized to data values that have a periodic, or regular, rate. Because duration in some way is tied to
our audible perception of the sound, it is important to note that sounds have an amplitude envelope, or
shape, which changes how we might perceive the length and character of the sound. A snare sound, for
example, is a short sound. The sound is loud for a short period of time but quickly dies off.

Envelope

Sounds change over time. An envelope refers to a shape that describes the various changes in sound
over time. A sound has an amplitude envelope, which describes the amplitude changes over time. A
sound has a frequency envelope, which describes frequency changes over time. A sound has a spectral
envelope, which describes frequency-amplitude changes together. Using envelopes—types of time-
variant controls—are useful in sonification. For example, an envelope generator is a common non-
periodic shape that can modify a sound and is made up of several segments. An envelope can have up to
four segments—attack, decay, sustain, and release— and each is shaped with "different time and level
values" (Stolet 2009).