Skip to content

Timbre tells the brain what arrived

Pitch says which note. Timbre says what is playing it, and listeners hear it in a fraction of a second. Choose the sound before you polish the melody.

Updated · 7 min read Lesson 1 of 10 in Sound Design

In short

  1. Timbre is what tells two sounds apart at the same pitch and loudness, and listeners pick it up within a fraction of a second.
  2. It comes from the balance of harmonics and from how a sound starts and changes, so the attack matters as much as the tone.
  3. Choose the sound before you polish the notes, and give parts that share a register clearly different colours.
In this lesson (8)

Same notes, different instrument

You program a melody and play it on a synth lead. Then a singer sings the same notes at the same tempo. You know at once which is which, before the first note has finished. Filter the synth or distort the vocal and you still do not confuse them.

The notes did not tell you that. The did: the quality that lets you tell two sounds apart at the same pitch and loudness. It works fast. In one study, listeners heard clips of popular recordings only 100 or 200 milliseconds long and still matched them to the right title and artist more often than chance. When the clips were played backwards or had their high frequencies removed, the shortest clips dropped to chance (Schellenberg, Iverson and McKinnon, 1999). A tenth of a second is far too short for a melody. It is long enough for a sound.

Why it matters

The sound is part of the part

When you choose a preset after the notes are written, you treat the sound as packaging. The listener hears it the other way round. The same four bars on a bright saw lead, a soft sine pluck and a slow pad are three different parts. One pushes, one sits close, one floats behind.

Timbre also decides whether parts stay separate in a mix. The ear sorts a mix into streams, one per source, and sounds with similar timbre in the same register tend to be grouped together (Bregman, 1990). Two similar pads in the same octave blend into one sound, and the quieter one is masked. Choose contrasting sounds from the start, one bright and edgy, the other round and soft, and they sit apart with little .

Figure 1. Three waves with the same period, so the same pitch. They differ only in which harmonics they contain and how strong each one is. Each is built from its first few harmonics, which is why the edges ripple.

The science

Harmonics and how they change

A pitched sound repeats. Anything that repeats can be built from sine waves at whole-number multiples of its repetition rate, the f0f_0. Its spectrum is a set of lines:

X(f)=∑n=1NAn δ(f−nf0)X(f) = \sum_{n=1}^{N} A_n \, \delta(f - n f_0)

Each term is one harmonic: a line at nf0n f_0 with amplitude AnA_n. The fundamental sets the pitch. The amplitudes set the spectral envelope, the contour across the tops of the lines, and that contour is a large part of what you hear as colour. A saw has every harmonic, falling as 1/n1/n. A square has only the odd ones, also falling as 1/n1/n. A triangle has only odd ones falling as 1/n21/n^2, which is why it sounds close to a sine.

Figure 2. The same 110 Hz note with two spectral envelopes. The lines sit at the same frequencies, so the pitch does not change. In the second the tops fall faster, which you hear as a rounder, darker sound.

Listen

Shape a sound with a filter

Chords and soft noise, or a real mix, through one filter, with the spectrum drawn live. Sweep the cutoff and raise the resonance.

Loading demo…
100%

Start with your speakers or headphones low.

The spectrum is only half of timbre. When listeners rate how different pairs of instrument sounds are, the same few properties keep explaining their judgments: how fast the sound rises at the start, where the centre of its spectrum sits, and how much the spectrum changes over time (Grey, 1977; McAdams and colleagues, 1995). The start matters for recognition too. Instruments become harder to identify once the starts of their notes are cut off (Saldanha and Corso, 1964). A sound's identity lives in how it starts as much as in how it sustains.

Try it

The four-sound test

Hear how much of a part's character comes from the sound rather than the notes.

0 of 7 steps done

The notes never change, yet you hear four different parts. The onset and the slope of the did more to the character than any note edit could.

Common mistake

Stacking the same colour

A weak line tempts you to add layers that sound like it, such as three saw leads in the same octave. Similar sounds that start together fuse into one, so the stack gets louder without getting more distinct, and the small detail that made one of them interesting is masked by the other two. Stacking near-identical waves also causes problems of its own, covered in the lesson on layered sounds.

The other mistake is choosing a sound in solo. A pad that sounds rich on its own may be the reason the vocal disappears. Judge timbre against the parts it has to live with.

Takeaway

Choose character before detail

Pick the sound before you polish the notes. Give each part its own face: if the lead is bright and sharp, keep the support round and slower. If two parts must share a register, make their onsets or their harmonic slopes clearly different. When a part does not work, try a different sound before you rewrite it.

3 questions

Check yourself

  1. Two tones have the same pitch and the same loudness, but they sound different. What differs?
  2. A saw and a square play the same 110 Hz note. Which statement is right?
  3. Listeners can match a 200 ms clip of a popular song to its title more often than chance. What are they using?

Sources

Show all 5
  • Bregman, A. S. (1990). Auditory Scene Analysis: The Perceptual Organization of Sound. MIT Press.
  • Grey, J. M. (1977). Multidimensional perceptual scaling of musical timbres. Journal of the Acoustical Society of America, 61(5), 1270-1277.
  • McAdams, S., Winsberg, S., Donnadieu, S., De Soete, G., & Krimphoff, J. (1995). Perceptual scaling of synthesized musical timbres: Common dimensions, specificities, and latent subject classes. Psychological Research, 58(3), 177-192.
  • Saldanha, E. L., & Corso, J. F. (1964). Timbre cues and the identification of musical instruments. Journal of the Acoustical Society of America, 36(11), 2021-2026.
  • Schellenberg, E. G., Iverson, P., & McKinnon, M. C. (1999). Name that tune: Identifying popular recordings from brief excerpts. Psychonomic Bulletin & Review, 6(4), 641-646.

Next lesson · 2 of 10

Warm, dark and dull are different soundsWarm adds body, dark tilts the top down, dull loses the attack. Treat all three as a low-pass request and you blur the part. Match the move to the word.
Portrait of Virzy Guns

Written by Virzy Guns

Songwriter and producer, founder of Virzy Guns Production. Now building HealingWave.