Describing timbre to other people can sometimes feel messy and vague. A certain amount of messiness is probably necessary for something so complex, but there are certain common concepts, terms, and approaches that can make communication about timbre more effective. We will introduce two important elements that affect how we perceive timbre: how it changes over time, and its component frequencies. We will then share two approaches to describing these.

The first major determinant of timbre is how a sound changes over time. For example, if you pluck a guitar string, the sound starts with a sharp, loud noise from that pluck, followed by a much quieter sustained note that slowly gets quieter until you can’t hear it any more. For many sounds, it’s useful to focus on two phases. The first is the “attack,” “onset,” or “articulation.” This is most obvious in instruments with naturally percussive attacks like drums, cymbals, piano, and plucked strings or when a singer starts with a consonant, but even wind instruments and bowed strings often have a slight change from their initial sound to what follows. The second phase is the “sustain” or “steady state.” This is a more consistent sound, though it may still change somewhat over time as a performer changes their vibrato or volume or as the instrument naturally gets quieter over time. If you practiced imitating timbres with your voice, it may be useful to think of representing the attack with your opening consonant (if any) and the sustain with your choice of vowel.

The second major determinant of timbre is its components. Even when we think we are only hearing a single pitch, most constant sounds are made up of more than one pitch or frequency. (For a fuller explanation, see here.) Timbre is strongly affected by which frequencies are present and how loud they are. This aspect of timbre is often visualized with a graph called a spectogram.

Describing Timbre 1: General Approaches

Descriptions of timbre often fit in a few general categories:

• Matter/Material. These descriptions focus on a material that we imagine was involved in making the sound. For example, we might describe a sound as brassy, metallic, or reedy.
• Action. Sometimes we focus instead on an action we might associate with a sound. Sometimes we are describing some aspect of the action we imagine a performer is doing to make the sound (breathy, strained), and sometimes we are describing an action we imagine the sound is doing either to the space around it or to the listener (piercing, biting).
• Analogies to other senses. Some terms describe sound by analogy to something we understand in another sense like sight or touch. Examples include brilliant, dull, dark, clear, warm, rough, and sweet.
• Onomatopoeia. We sometimes use words that sound like what they mean, like buzzing, rattling, hissing, or clanging.
• Other analogies. Because timbre is hard to describe, analogies in general can be useful. Common examples include rich, thick, pure, round, deep, and heavy.

Describing Timbre 2: Binary Oppositions/Spectra

Some descriptive words for timbre pair nicely with their opposites. This is particularly helpful because it allows us to compare two sounds and, for example, describe one as “darker” and the other as “lighter.” Sometimes a sound will fit clearly with one term and not its opposite, but these are really on a spectrum and sounds in the middle of the spectrum may be hard to describe. We also perceive sounds in relation to their context, and an instrument that sounds dark in one context may sound light in another. Still, these terms are commonly used and facilitate comparisons.

The list below is drawn primarily from the work of Megan Lavengood (who in turn drew on a book by Robert Cogan), focused on the terms we think are most common. Many of the terms have fairly intuitive meanings; we have annotated a few where we think it’s helpful. Fuller technical definitions are in Lavengood’s article. Each pair is linked to Lavengood’s audio example, each of which is also illustrated with a spectogram.

• Bright/Dark
• Pure/Noisy
• Full/Hollow
• Percussive/Legato: this is primarily influenced by a sound’s attack
• Harmonic/Inharmonic: we can understand this intuitively by how much dissonance or clash there is in a timbre; bells, for example, often have a complex, dissonant (“inharmonic”) sound
• Steady/Wavering: Lavengood does not have an audio example but relates this primarily to the presence or absence of vibrato