Understanding the voice (4/4)

The voice timbre

The sound produced by the vocal cords is formed, like any vibration, of a mixture of frequencies, called harmonics, which are multiple of a basis frequency called fundamental. The distribution of the magnitude of these frequencies is called spectrum in acoustic or timbre in music.

The spectrum displayed by Canta

Canta shows the lower part of the spectrum of the voice, in the region where lies the fundamental, between 80 and 1500 Hz.

The red line represents the position of the fundamental, the other bumps are the different harmonics.

Our ear is able to detect not only the fundamental of a sound, but also recognize different spectrums.

The voice resonators

The voice resonators

The spectrum of the sound produced by our vocal chords is changed by the resonances of the different harmonics in the cavities located between the vocal cords and the outside: the pharynx, the mouth and the nasal cavities. Depending on the form of these cavities, certain frequencies will be faded while others will be amplified and will be more audible. The areas where the harmonic frequencies are amplified are called formants.

The biggest changes in the spectrum are caused by the position of the tongue, the jaw and the opening of the mouth. One can alter the shape of the oral cavity and produce different vowels with the same pitch.

The detection of the first formants allows us to distinguish between vowels and then allow us to speak (and to sing). What counts for understanding of speech, is the relative position of the formants and not so much their absolute value, thus it is possible to sing vowels at different pitch and to remain understandable.

In the diagram below, you can see how the tongue and the lips can alter the shape of the oral cavity and thus change the spectrum of the produced sound.

The buccal cavity changes
The vowels

The nasal vowels are produced by lowering the soft palate that causes a reduction of the high frequencies.

The consonants are not harmonic sounds, they are transients without fundamental frequency. They are produced by different modes of opening and closing of the mouth.

Generally we do not confuse two people saying the same vowel. This stems from the rest of the spectrum, including the highest part which is not involved in the detection of vowels, but which is fundamental for the voice timbre. The color of the voice is unique to each individual, it depends on its anatomy and how it articulates the words.

The singing formant

It is an area of resonance of the human voice located around 2800/3000 Hz for the men and 3000/3200 Hz for the women. This is the region where our ear is the most sensitive, while the instruments of classical music only produce low frequency in this area, allowing the voice going out over the orchestra.

The diphonic singing

It is a form of singing used in Central Asia in which one selects some harmonics that become so intense that it gives the impression of a second melody. We can test this approach by dividing the oral cavity with the tip of the tongue.

Normal spectrumDiphonic spectrum
Spectrum of a normal "o"
Spectrum of a diphonic "o"

Back home

Previous: (3) The pitch control


Freeware for Windows XP and Vista.

Audio card and external microphone required

857 Ko

Learn to sing in tune with CANTA

Canta to learn to sing in tune


Chaumet SoftwareCanta - a tool to sing betterLearn to sing in tune

Canta is the perfect teacher to get singing lessons while having fun singing your favorite songs.

With Canta you will learn how to sing, you will learn to sing better.