Filtering
Any part of a sound spectrum can be turned up or turned down.
This is known as filtering, and it can be done in a variety of ways (e.g. playing music behind a curtain or listening to music through a tube held up to your ear). The most common method, however, is to use electronic tools.
This is known as filtering, and it can be done in a variety of ways (e.g. playing music behind a curtain or listening to music through a tube held up to your ear). The most common method, however, is to use electronic tools.
Filtering terms
- Q value refers to the emphasis put on the range to be filtered; a large Q value sharpens (narrow band), and a small Q value softens (wide band)
- Cutoff frequency is the frequency starting at which the spectrum is being changed
- Slope refers to how steep the filtering graph is (e.g. 6 – 12 – 24 dB, from sboothest to sharpest)
Here are some of the most common filtering protocols (with practical examples):
- low pass: allowing partials under the cutoff frequency to pass and attenuating higher ones (ex. remove hiss)
- high pass: allowing partials over the cutoff frequency to pass and attenuating lower ones (ex. remove rumble)
- band pass: allowing partials in a specific frequency range to pass and attenuating higher and lower ones (ex. kick drum emphasis)
- notch (band reject): attenuating partials only in a specific frequency range and allowing all others to pass (ex. correct nasal vocals or remove a background hum)
The following images illustrate types of filtering.
vaimennus = attenuation
taajuus = frequency
päästökaista = passing band
estokaista = reject band
vaimennus = attenuation
taajuus = frequency
päästökaista = passing band
estokaista = reject band
Formant
A formant is a kind of array of fixed filters, with several attenuating and amplifying filters or bands. The body of a violin or a flute and the vowels of a singer each have their own formant.
A formant changes the timbre of a sound passing through it. The body of a violin resonates in certain frequency bands, amplifying partials in those bands and attenuating others.
Vowel formants are created by shaping the vocal tract to change the timbre of the sound produced by the vocal cords.
Formants have the effect for instance of the vowel A always sounding like the vowel A regardless of the pitch at which it is sung.
Throat singing involves producing a low, edgy fundamental tone and then using the mouth cavity as a narrow band pass filter to isolate various partials in the steady tone. This can be damaging for the larynx unless done properly, so experiment with care!
A formant changes the timbre of a sound passing through it. The body of a violin resonates in certain frequency bands, amplifying partials in those bands and attenuating others.
Vowel formants are created by shaping the vocal tract to change the timbre of the sound produced by the vocal cords.
Formants have the effect for instance of the vowel A always sounding like the vowel A regardless of the pitch at which it is sung.
Throat singing involves producing a low, edgy fundamental tone and then using the mouth cavity as a narrow band pass filter to isolate various partials in the steady tone. This can be damaging for the larynx unless done properly, so experiment with care!
Here are illustrations of some vowel formants.
By filtering a tone as shown in the image, we can make it sound like the vowel A, I or U.
Voimakkuus = amplitude
a-vokaali = vowel a
i-vokaali = vowel i
u-vokaali = vowel u
Benaden mukaan = after Benade
By filtering a tone as shown in the image, we can make it sound like the vowel A, I or U.
Voimakkuus = amplitude
a-vokaali = vowel a
i-vokaali = vowel i
u-vokaali = vowel u
Benaden mukaan = after Benade
A vocoder is an active filter array used in synthesisers. It analyses a sound source and adapts the dynamic filter array to change the other sound passing through to resemble the sound source.
An early version of this was the VoiceBox used by guitarist Jeff Beck. It conveyed his guitar sound to his mouth through a small speaker, allowing him to ‘speak’ guitar sounds through his mouth, which is quite a fun sound.
(Jeff Beck, 1992, Full Moon Boogie, with Jan Hammer Group Live)
An early version of this was the VoiceBox used by guitarist Jeff Beck. It conveyed his guitar sound to his mouth through a small speaker, allowing him to ‘speak’ guitar sounds through his mouth, which is quite a fun sound.
(Jeff Beck, 1992, Full Moon Boogie, with Jan Hammer Group Live)
The adjacent figure shows vowels and their frequency peaks. The fundamental frequency at which the vowels were spoken is at the bottom, and the most prominent formant bands show up in red.
Lower down, there is a major scale that I sang on the vowel A.
The figure shows that although the fundamental frequency changes, certain bands remain constant, because my mouth cavity is shaped for the vowel A.
The figure shows that although the fundamental frequency changes, certain bands remain constant, because my mouth cavity is shaped for the vowel A.
Facts:
- Filtering changes timbre
- Filtering is used to correct or emphasise certain partials in a sound
- Filtering parameters include Q value, cutoff frequency and slope
- A formant is an array of fixed filters that shapes sound conveyed through it
version 29.6.2022