How Does the Human Voice Work? Vocal Folds, Resonance and Range

The vocal folds are two bands of muscle covered in a loose mucous membrane, stretched across the airway inside the larynx, the box of cartilage behind the Adam's apple. When a person breathes they are held apart; to speak they are brought together, and the air pushed up from the lungs forces them apart, the pressure drops as the air rushes through, and they snap shut again, in a cycle that repeats at the frequency of the note being sung, about 100 times a second for a man's speaking voice, 200 for a woman's, over 1,000 for a soprano's top notes. The result is a buzz, rich in harmonics, and not yet a vowel. Pitch is raised by tightening the folds, using muscles that stretch them as a guitarist stretches a string, and loudness by pushing more air through them, which is why shouting is work for the breath.

The buzz passes up through the throat, mouth and nose, and that column of air resonates at frequencies set by its shape, reinforcing some harmonics of the buzz and damping others. Move the tongue, jaw and lips and the shape changes, and the pattern of reinforced frequencies, the formants, changes with it; the ear reads the pattern as a vowel. The same buzz shaped by a wide open mouth is ah and by a mouth nearly closed with the tongue high is ee, and the difference between one speaker's ah and another's is the difference in the sizes and shapes of their tubes. The parts and what they do:

• •The lungs and diaphragm: the power supply, whose control is what singers call breath support
• •The vocal folds: the source, setting pitch by their tension and length, and producing a whisper when they are held apart and air rushes through
• •The throat, mouth and nose: the resonator, shaping the buzz into vowels and giving the voice its timbre
• •The tongue, lips, teeth and palate: the articulators, which interrupt the flow to make consonants
• •The false folds, the epiglottis and the ventricles: structures whose adjustment gives the growl, the twang and the operatic ring

Men's folds are longer and thicker than women's, since testosterone at puberty grows the larynx, which is why the voice breaks and why men's voices are about an octave lower; taller people have longer vocal tracts and lower formants, which is why a large man sounds large. The rest is the fine structure: the exact length and mass of the folds, the shape of the nasal cavity, the habits of a lifetime of speaking in one language and one accent, and the small irregularities in the folds' vibration that give a voice its grain. A voice is as identifiable as a face and for the same reason, that its features are many and their combination is unique, and it changes with age as the folds stiffen and thin, which is why the elderly sound as they do and why a singer's career has a shape.

A trained voice is the same instrument used more efficiently. Singers learn to manage the breath so that pressure at the folds stays steady across a phrase; to keep the folds vibrating cleanly through the break between the lower and upper registers, the chest and head voices, where the muscles that set pitch hand over from one to another; and to shape the resonator so that the harmonics of the voice fall where the ear is most sensitive, around 3,000 hertz, which produces the ring that carries an unamplified voice over an orchestra. Range is partly given and partly extended by training, and the classification of voices, from bass through baritone and tenor for men and contralto, mezzo and soprano for women, is by range and weight together. The extremes are wide: a bass may sing down to 65 hertz and a coloratura soprano up past 1,500, and the folds that do both are a couple of centimetres of tissue.

The folds are delicate, and a voice is lost by using them badly. Shouting, singing without support, smoking and reflux swell and thicken them, and the nodules and polyps of the overworked voice, the singer's nodes that end careers, are calluses on the folds' edges from being slammed together too hard for too long; rest, hydration and retraining cure most of them, and surgery some. The voice also carries the state of the body, which is why a cold, tiredness, fear and grief are audible, and why the stress of a phone call can be heard by the person at the other end; the instrument sits in the throat, between the breath and the brain, and everything that passes through either shows in it.

For a long time nobody could watch a living voice being made, because the vocal folds sit out of sight in the throat. In the 1850s the singing teacher Manuel Garcia used a small dental mirror held at the back of his mouth, a second mirror and sunlight to look down at his own larynx while he sang, and he published what he saw. Doctors adopted the method, and the laryngeal mirror became a standard tool for examining the voice box. It showed the folds opening and closing, but not how they moved during each tiny cycle.

The folds vibrate far too fast for the eye, so modern clinics use tricks. In videostroboscopy a flashing light is timed slightly out of step with the voice, so that each flash catches the folds at a slightly later point in their cycle, and the picture seems to show them moving in slow motion. High-speed cameras can now film thousands of frames a second directly. These views let specialists spot a nodule, a stiff patch of tissue or folds that do not meet evenly, and they have taught researchers how the wave of movement ripples across the fold surface.

Most people dislike hearing a recording of their own voice, and the reason is physical. When you speak, the sound reaches your inner ear by two routes: through the air around your head, like everyone else's voice, and directly through the bones of your skull, which carry the vibrations of the throat and mouth. Bone conduction favours the lower frequencies, so the voice you hear while talking is fuller and deeper than the one others hear. A recording captures only the airborne sound, so it seems thinner and higher, but it is much closer to how you sound to everyone else.

Helium shows the difference between the source and the filter. Breathe it and your voice turns squeaky, which is usually explained as helium raising the pitch. It does not, or only slightly. The vocal folds still vibrate at much the same rate, but sound travels nearly three times faster in helium than in air, so the resonances of the throat and mouth move to higher frequencies. The buzz keeps its pitch while the formants that colour it shift upwards, producing the cartoon timbre. Deep-sea divers who breathe helium mixtures have the same problem, and electronic unscramblers are used to make their speech understandable.

Every voiced sound contains a fundamental pitch plus a series of fainter harmonics above it, and the shape of the mouth and throat decides which harmonics are strengthened. In Tuvan throat singing, from the Tuva region of southern Siberia, singers shape their vocal tract so precisely that one high harmonic becomes loud enough to be heard as a separate whistling melody above a steady low drone.

The singer is not using two sets of vocal folds; the second note is an existing harmonic selected and amplified by resonance, the same principle that makes vowels sound different, pushed to an extreme. Similar overtone singing traditions exist in Mongolia and elsewhere. Hearing it is one of the clearest demonstrations that a single voice always contains many pitches at once.

The human voice is a buzz produced by the vocal folds in the larynx snapping open and shut in the air from the lungs, at a frequency that sets the pitch, filtered into vowels by the resonance of the throat and mouth and cut into consonants by the tongue, lips and teeth. Voices differ by the size and shape of the folds and the tract, and singers train the breath, the passage between registers and the resonance that makes a voice ring, on an instrument that shouting, smoking and strain can damage.