The Colours of Noise
To most people, noise is generally described by its loudness or pitch. But did you know that noise can also be differentiated by colours?
Some of you may already know about white noise, which is typically said to be the static noise generated by air conditioners (although technically, the AC’s hum is not exactly white noise). The sounds that we experience daily consist of various colours: pink, brown, green, blue etc. In fact, different noise colours have their own unique properties, each of them having different “abilities”, such as being used for music production, acoustics testing or promoting relaxation. (Neal, 2016) (Geere, 2011)
Why are they different?
The noise colours can be differentiated by their spectral density, which is the distribution of signal power (amplitude) over a range of frequencies. Noise with different spreads will be given different colour names, where they are named based on a slight relation to the analogy of the colours of light. For example, white noise consists of all the audible frequencies, which is similar to white light that includes all the frequencies in visible range.
The way the frequencies are distributed can greatly affect the sound being produced. Frequencies that are spaced at intervals (i.e. musical sound waves) are capable of forming a harmonic structure that gives a special tone quality to sound. This is the reason why we find musical sound generally pleasant to our hearing. Conversely, the so-called “noises” that we may find annoying, like boots stomping or car honks, are formed by sporadic waveforms, which is a random distribution of frequency and amplitude.
But there is one more category: the coloured noises. Although they are continuous signals (not random), they can sometimes be unpleasant, hence the term “noise”. “Noise” was from a Latin word which means nausea, where in audio engineering, it also means any unwanted information disturbing or interfering the signal. One common example for this would be the static noise on the radio.
White noise can be described as the hissy noise (static) usually produced by TV or radio, or you can think of it as the sound of waves by the beach. Ultimately, it is a noise that is capable of overriding background noise. White noise is a mixture of all the frequencies that are audible to humans (20 Hz – 20 kHz), released randomly with equal power, which means its frequency spectrum is totally flat. A quick way to think about it is that, the signal band between 20 – 40 Hz holds the same amount of power as the band from 2000 – 2020 Hz. This is analogically similar to white light, in which every band of the spectrum shines at the same brightness, combined to become white.
Having the ability to override background noise, white noise can be used as part of siren on emergency vehicles because it is easy to locate the source of the sound. Due to its equal amount of high to low frequencies, white noise is applicable in sound-masking tools, for instance those used for tinnitus patients, or purely to help improve sleep quality.
Audio Example: White Noise
Unlike white noise, pink noise has spectrum that drops logarithmically as time passes., with equal power in bands that are proportionally wide rather than or equal width. The signal in 20 – 40 Hz has the same power as compared to the signal band 2000 – 4000 Hz. Therefore, we can say that pink noise works similarly with the humans’ auditory system, because we hear in octaves (doubling of frequency band). This gives the effect where humans will not find pink noise to be too harsh compared to white noise.
The sound energy in pink noise falls out as frequency increases, in fact, dropping by half as frequency doubles. So, every octave has equal power giving a more balanced sound. Pink noise is very useful especially to be applied as a signal to test amplifiers or loudspeakers. It is sometimes also seen in meteorological data series or in the radiation-output of some of the astronomical bodies.
Recently, pink noise is greatly used as the alternative to white noise for improving sleep or concentration. Studies have shown that pink noise helped participants to achieve deeper sleep, and some health blogs also claimed that it is the key to a better night’s rest.
Brown (Brownian) Noise
Note: It is not the same as Brown Note!
The brown noise, sometimes known as Brownian noise, is actually a deeper version of pink (not brown!). The reason that it is named this way is because, the signal of the brown noise mimics the Brownian motion, which is like a random walk pattern / random particle movement in liquid. It sounds deeper and bassy, which can be thought of as the low-roaring sound of heavy winds or ocean waves.
Audio Example: Pink Noise
Audio Example: Brown Noise
Now that we have introduced the red and brown noises that are on the front end of the spectrum (more concentrated at the lower frequency range), we will now proceed to the other end of the spectrum. One of them is the blue noise, that sounds like the hissing-sound of water spray. It can be understood as the inverse of pink noise. Blue noise changes proportionally with increasing frequency, but unlike pink noise, the blue noise increases with frequency rather than decreasing.
Sound engineers tend to use blue noise for audio dithering, which means the process of adding noise to a signal to minimize distortions that are present during production. This is because the blue noise has high-pitched noise that is harder for the listener to detect. Adding these noise signals randomizes the errors, hence smoothing out the rough edges.
Similar to blue noise, violet noise also has more energy concentration at the higher end of the spectrum. The violet noise is the inverted version of Brownian noise, where its power gets denser per octave as the frequency increases over a finite frequency range. Its high frequencies can mask off the high frequencies that affects tinnitus sufferers, hence it is also used for tinnitus treatment.
Grey noise is different from those explained before – it has high power at both ends of the frequency spectrum, but very little centred at the range of human hearing. Which means, grey noise typically sounds the same at every frequency. In fact, it is actually white noise, but tuned to the psychoacoustics of average human ears specifically. Therefore, grey noise is very useful in studies related to hearing difficulties, where researchers use it to assess how much a person’s hearing differs from the average standard.
In general, only white, pink and blue noise are given official definitions in the federal telecommunications standard. Brown and grey are accepted in certain industries. As for the other colours of the noise rainbow, they are only informally defined. For example, orange noise consists of a spectrum having a few bands being eliminated fully, so it only plays everything that is not perfectly in tune and does not play anything that is in tune! This is why it is given “orange” as its name, to represent its “sourness” in the noise. Violet noise that we mentioned in front, is simply a more intense version of blue noise, where the concentration of energy is even higher in the highest audible frequencies.
One last noise to be introduced, that is given an official meaning, is the black noise. It is a spectral density that is approximately zero power at every frequency. Some defines it as pure silence, where others say that it is the output of an active noise control system the cancels out the existing noise (negative noise). In a way, it is the complete opposite of white noise.
And there you go, the colours of noise. Out of all that was being mentioned above, which one did you find the most interesting? Leave your thoughts in the comments section below!
Geere, D. (2011, April 07). White, pink, blue and violet: The colours of noise. Retrieved November 30, 2020, from Wired: https://www.wired.co.uk/article/colours-of-noise
Neal, M. (2016, February 16). The Many Colors of Sound. Retrieved November 30, 2020, from The Atlantic: https://www.theatlantic.com/science/archive/2016/02/white-noise-sound-colors/462972/
Written by Khei Yinn Seow
Posted on December 3, 2020