Voice Characteristics: What They Are and How They Work

Voice characteristics are the measurable acoustic properties that make each person’s voice distinct, and they arise from a surprisingly physical process: air from the lungs pushes through vibrating folds of tissue in the larynx, producing a raw buzzing sound that is then shaped by the throat, mouth, and nasal passages into what we recognize as speech or song. The interplay between those vibrating folds and the resonating cavities above them determines everything from your pitch and loudness to the subtle tonal quality that lets a friend recognize your voice on the phone in under a second. What makes the system remarkable is how much complexity emerges from a relatively simple mechanical starting point.

How the Vocal Folds Actually Vibrate

The engine of the voice is a pair of small, layered tissue structures in the larynx called the vocal folds. They do not vibrate because the brain sends rhythmic signals telling them to open and close hundreds of times per second. Instead, they oscillate passively, driven by the air pressure coming up from the lungs. The framework explaining this is sometimes called the myoelastic-aerodynamic theory, and it has been the dominant model in voice science for decades. In plain terms, the elasticity of the vocal fold tissue (the “myoelastic” part) and the forces of the airstream passing through them (the “aerodynamic” part) combine to create a self-sustaining cycle of opening and closing.

The key to sustaining that cycle is an asymmetry in shape. As the folds open, the gap between them takes on a convergent shape (wider at the bottom, narrower at the top), and subglottal pressure from below pushes them apart. As the folds close, the gap becomes divergent (narrower at the bottom, wider at the top), and pressure drops sharply, pulling them back together. This switching between convergent and divergent configurations is what keeps the vibration going, because the aerodynamic forces are stronger during opening than during closing.

A further refinement involves what researchers call “vertical phase differences” or mucosal waves. The bottom edge of each vocal fold moves slightly ahead of the top edge, creating a ripple-like motion along the fold’s surface. This ripple turns out to be a highly efficient way to transfer aerodynamic energy into tissue vibration, more so than simple pressure oscillations from below.

² In chest voice, these vertical and horizontal phase differences are large and pronounced, involving substantial tissue displacement. In falsetto, the tissue moves more uniformly, and the driving asymmetry comes mainly from the inertia of air moving through the narrow glottal slit.³

How the Vocal Tract Shapes Raw Sound Into Recognizable Voice

The buzz produced at the vocal folds is only the raw material. By itself, it sounds nothing like speech or singing. What transforms it is the vocal tract: the pharynx (throat), the oral cavity, and the nasal passages. These cavities act as an acoustic filter, amplifying certain frequencies and dampening others depending on their shape at any given moment. The frequencies that get amplified are called formants, and they are largely what allow you to distinguish one vowel from another, and one speaker from another.

⁴

The resonances of the vocal tract determine the spectral envelope of the voice: the overall shape of which frequencies are loud and which are quiet. They also contribute to timbre, loudness, and efficiency of sound production. In speech, these resonances carry the phonemic information that lets listeners decode words. In singing, they contribute to the tonal color that distinguishes, say, a warm operatic baritone from a bright pop tenor.

⁵

Computational modeling has shown that the vocal tract and the vocal folds are not entirely independent systems. Adding even a simple uniform tube above the folds changes the voice source signal. Extreme constriction at any point along the tract reduces both the average and the peak-to-peak flow of air through the glottis. One interesting exception: narrowing the epilarynx (the small tube just above the vocal folds) can increase the sharpness of the airflow cutoff during each cycle, which boosts high-frequency energy. This is relevant to certain singing techniques that exploit epilaryngeal narrowing to produce a more “ringy” or projected sound.

⁶

Pitch, Loudness, and the Tiny Variations You Never Notice

The most obvious voice characteristic is pitch, which corresponds to the fundamental frequency of vocal fold vibration. Thinner, tighter, and longer vocal folds vibrate faster, producing a higher pitch. Thicker, looser, and shorter folds vibrate more slowly. The vocal folds adjust their tension primarily through the action of laryngeal muscles that tilt and slide the cartilages to which the folds are attached. Because the different tissue layers of the vocal folds are all connected at the same anchor points, they cannot be lengthened independently, so the stress within the tissue fibers becomes the critical variable for controlling pitch.

⁷

Loudness, meanwhile, is primarily a function of how forcefully air is pushed through the glottis and how completely the folds close during each cycle. Greater subglottal pressure means a larger volume of air displaced per cycle, which translates to a louder sound.

Beyond pitch and loudness, voice scientists measure two subtle characteristics that reflect the stability of vocal fold vibration. Jitter is the tiny cycle-to-cycle variation in fundamental frequency, and shimmer is the corresponding cycle-to-cycle variation in amplitude. You cannot consciously hear these micro-fluctuations in a healthy voice, but acoustic analysis software can detect them, and they serve as clinical markers when they become excessive. Research with male speakers producing sustained vowels at soft, moderate, and loud levels found that both jitter and shimmer decrease as loudness increases, meaning a louder voice tends to be a more stable voice, at least in terms of these micro-perturbations.

⁸

What Makes a “Bright” or “Dark” Voice

Timbre, the tonal quality that makes two voices singing the same note at the same volume still sound different, is shaped in large part by the vocal tract. Opera singers, who have some of the most refined control over timbre, offer a useful window into the mechanics. An MRI-based study of opera singers producing the vowel /a/ in bright versus dark voice qualities found that the difference comes down to the lower pharynx. A bright voice was associated with a narrower lower pharynx, which raised the first and second formant frequencies. A dark voice came from a wider lower pharynx, which lowered those same formants.

⁹

This finding illustrates a broader principle: much of what we perceive as voice “color” is not about the vocal folds themselves but about the resonating chambers above them. Singers and actors learn to manipulate these chambers, often without understanding the acoustics explicitly. They talk about “placing” the voice forward or back, or “opening the throat,” and these descriptions map onto real changes in pharyngeal and oral cavity dimensions that shift formant frequencies in predictable ways.

Why Voices Differ Between Sexes and Across the Lifespan

The most dramatic natural variation in voice characteristics happens during puberty, especially in males. Rising testosterone levels drive the growth of the larynx and the lengthening and thickening of the vocal folds. A study tracking adolescents found that in males, changes in vocal tract anatomy mediated the relationship between bioavailable testosterone and the resulting acoustic changes in the voice.

¹⁰ The adult male larynx is on average about 40 percent larger than the adult female larynx, which is why the typical male speaking pitch sits roughly an octave lower than the typical female speaking pitch.

At the other end of life, the voice changes again. The vocal folds lose some of their elasticity and mass as the extracellular matrix that gives them their layered, pliable structure degrades. The fibrous proteins and glycosaminoglycans within the folds change in density and spatial arrangement, altering the biomechanical properties that allow smooth vibration. These age-related changes, sometimes called presbyphonia, often produce hoarseness, breathiness, and a narrowing of pitch range. Systemic factors like hormonal shifts and reduced lung capacity also play a role, making age-related voice changes difficult to treat with any single intervention.

¹¹

Vocal Registers and the Physics of Switching Between Them

Most people are intuitively familiar with the experience of shifting between a heavier, richer lower voice and a lighter, airier upper voice. Voice scientists describe these as different registers, with “chest voice” (or modal register) and “falsetto” being the most commonly discussed pair. The difference between them is largely about how the vocal folds vibrate. In chest voice, the folds make full contact along their depth and length, with prominent mucosal waves. In falsetto, the folds are stretched thinner and vibrate with less contact and more uniform motion.

The transition between registers is not always smooth. Computational modeling and empirical data show that intraglottal pressures can change abruptly when the glottal geometry shifts relatively gradually from convergent to divergent. The nearly rectangular shape associated with mixed registration, where the fold surfaces are almost parallel, is inherently less stable than the highly angled shapes of pure chest or pure falsetto. Stabilizing that intermediate shape requires either reducing the pressure across the glottis or carefully balancing the stiffness of the upper and lower tissue layers of the folds.¹² This is why register transitions often produce an audible “break” or crack in untrained singers: the system flips abruptly between two stable states rather than passing smoothly through the unstable middle ground.

How the Brain Processes Vocal Sounds

The human brain does not treat voice sounds the same way it treats other sounds. Functional MRI studies have identified regions along the superior temporal sulcus, sometimes called “temporal voice areas,” that respond more strongly to vocal sounds than to matched non-vocal sounds like environmental noises or musical instruments. About 94 percent of participants in one large imaging study showed bilateral patches of significantly greater response to vocal than to non-vocal sounds in these regions.

¹³

These voice-selective areas appear to handle different aspects of vocal information in different locations. Speech sounds elicit greater responses than non-speech vocalizations across most of the auditory cortex, including primary auditory areas, on both sides of the brain. But the right anterior superior temporal sulcus stands out: it responds more strongly to non-speech vocal sounds (like laughing, sighing, or crying) than to frequency-scrambled versions of the same sounds, even when those scrambled versions preserve the same spectral content. This suggests those right-hemisphere regions are specifically involved in extracting paralinguistic information from voices, things like emotional tone, speaker identity, and physical characteristics, rather than decoding words.

¹⁴

What Your Voice Tells Other People

Listeners extract a surprising amount of social information from voice characteristics, often without realizing it. Both fundamental frequency (pitch) and formant spacing (related to vocal tract length) influence how dominant and attractive a speaker is perceived to be. Lower pitch and closer formant spacing, both associated with a larger body, predict higher dominance ratings. In one study, men were rated as stronger, better fighters, and more socially dominant as their voices were manipulated to sound more masculine.

¹⁵

Attractiveness judgments follow a related but distinct pattern. Women’s ratings of male vocal attractiveness were predicted by low mean pitch, low formant dispersion, high intensity, and attractive word content. Interestingly, low formant dispersion was perceived as attractive only by women in the fertile phase of their menstrual cycle, suggesting that sensitivity to certain vocal cues may shift with hormonal state.

¹⁶

The dominance and attractiveness findings do not always align neatly. A very masculine voice reliably signals dominance, but the attractiveness boost is less straightforward. One study found that the apparent increase in attractiveness from a lower voice was driven mainly by listeners finding feminized vocal manipulations unattractive, rather than finding the most masculinized versions especially attractive.¹⁵ In other words, there may be more of a floor effect than a ceiling effect when it comes to vocal masculinity and appeal.

The Descended Larynx and Human Vocal Range

One of the most distinctive features of human vocal anatomy, compared to other primates, is the low position of the larynx in the throat. In most other primates, the larynx sits higher, which creates a shorter pharyngeal cavity. In humans, the larynx descends during early development, creating a two-tube vocal tract: the oral cavity that all primates share, and an enlarged pharyngeal cavity that is essentially unique to humans. This configuration, along with an agile tongue and rapid lip and jaw movements, gives humans far more articulatory latitude than any other primate. The dynamic changes to vocal tract resonances that this anatomy permits are exactly what allow us to produce the wide range of vowels and consonants that make up human languages.

¹⁷

The trade-off for this vocal flexibility is a well-known one: because the larynx sits lower, the pathways for food and air cross in the pharynx, making humans more vulnerable to choking than species with a higher larynx. That evolutionary gamble, if it can be called one, appears to have paid off handsomely in terms of communicative ability.

Involuntary Vocal Adjustments in Noisy Environments

Your voice does not just respond to what you want to say. It also responds automatically to the acoustic environment. The Lombard effect is the involuntary tendency to speak louder, at a higher pitch, and with altered formant characteristics when background noise increases. It is not a deliberate decision to “talk over” the noise; it happens even when speakers are unaware of the change.

¹⁸

A study measuring the Lombard effect across a range of ambient noise levels from 35 to 85 decibels found that the relationship is not linear. Speakers showed a modest increase in vocal output at lower noise levels, but above roughly 58 decibels, the rate of increase nearly doubled. The presence or absence of hearing loss did not significantly change the strength of the effect, suggesting the response is deeply embedded in the vocal motor system rather than being purely a function of auditory feedback.

¹⁹

How Language Experience Tunes Vocal Control

The feedback systems that control your voice are not just hardwired reflexes. They are shaped by the language you grew up speaking. A cross-language study compared speakers of Cantonese and Mandarin, both tonal languages, while their voice pitch was unexpectedly shifted through headphones during sustained vowel production. When the pitch shifts were large (200 to 500 cents), Cantonese speakers produced significantly smaller compensatory responses than Mandarin speakers. Cantonese speakers also showed a systematic decrease in response magnitude as the shift got larger, a pattern not seen in Mandarin speakers.

²⁰

The explanation likely relates to the different demands of the two tonal systems. Cantonese has six tones with relatively fine pitch distinctions, so its speakers may develop tighter constraints on how much they let external perturbations push their pitch around. Mandarin has four tones with larger pitch movements, potentially making its speakers more tolerant of large pitch shifts. The broader point is that even the most automatic, low-level aspects of vocal motor control are calibrated by years of language use.

Singing Training and Changes in Brain Representation

If language experience shapes vocal control, dedicated vocal training reshapes it further. Trained singers show more accurate modulation of larynx height when imitating changes in vocal tract length, as measured with MRI. This behavioral advantage is underpinned by stronger neural representation of vocal tract length within a region of right dorsal somatomotor cortex that has been previously linked to singing experience.

²¹

What makes this finding interesting is that the enhanced control was demonstrated not during singing, but during a speech imitation task. This suggests the neural and behavioral refinements that come from singing training are not confined to musical performance. They spill over into ordinary speech motor control, giving trained singers finer-grained command over vocal tract configuration even in non-musical contexts. The broader neural control network for singing involves a distributed set of brain regions whose connectivity and representation change with various lengths and types of training, a phenomenon that researchers continue to map.

²²

When Voice Characteristics Signal a Problem

Because voice characteristics depend on the precise biomechanical properties of the vocal folds and the surrounding structures, changes in those properties show up acoustically. Vocal fold nodules, benign lesions that develop from chronic mechanical stress on the fold mucosa, are a common example. In children with nodules, studies find elevated jitter, shimmer, and noise-to-harmonics ratio compared to children without voice problems, reflecting the irregular vibration patterns caused by the added mass and stiffness of the lesions. Maximum phonation time, a simple measure of how long someone can sustain a vowel on one breath, is also reduced.

²³

Acoustic analysis has become a valuable clinical tool precisely because these parameters are sensitive to structural changes that may not yet be visible on examination. A voice that sounds breathy or rough to the ear can be quantified with jitter and shimmer measurements, tracked over time, and compared before and after treatment. Inflammation markers in the blood can even complement the acoustic picture: children with vocal nodules showed higher neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios than controls, highlighting that what presents as a voice problem often involves systemic inflammatory processes alongside local mechanical damage.²³

1
The Journal of the Acoustical Society of America. Measuring intraglottal pressure of physical model of oscillating vocal folds
2
PubMed. Integrative Insights into the Myoelastic-Aerodynamic Theory and Acoustics of Phonation. Scientific Tribute to Donald G. Miller
3
PubMed. Comments on the myoelastic-aerodynamic theory of phonation
4
Oxford Research Encyclopedia of Linguistics. The Source–Filter Theory of Speech
5
PubMed Central. Vocal tract resonances in speech, singing, and playing musical instruments
6
PubMed Central. The influence of source-filter interaction on the voice source in a three-dimensional computational model of voice production
7
PubMed Central. Predicting Achievable Fundamental Frequency Ranges in Vocalization Across Species
8
Journal of Voice. Influence of mean sound pressure level on jitter and shimmer measures
9
The Journal of the Acoustical Society of America. Effects of vocal tract configuration on bright and dark timbres in opera singing
10
PubMed. Age- and sex-related variations in vocal-tract morphology and voice acoustics during adolescence
11
PubMed Central. Presbiphonya
12
PubMed Central. Bi-stable vocal fold adduction: a mechanism of modal-falsetto register shifts and mixed registration
13
PubMed Central. The human voice areas: Spatial organization and inter-individual variability in temporal and extra-temporal cortices
14
PubMed. Human temporal-lobe response to vocal sounds
15
Royal Society Open Science. Low voice pitch, when accompanied by video, increases perceived dominance but not attractiveness
16
PubMed Central. Different Vocal Parameters Predict Perceptions of Dominance and Attractiveness
17
Current Biology. Evolution of human vocal production
18
PubMed Central. The Lombard effect observed in speech produced by cochlear implant users in noisy environments: A naturalistic study
19
Scientific Reports. Lombard effect, intelligibility, ambient noise, and willingness to spend time and money in a restaurant amongst older adults
20
PubMed Central. Effect of tonal native language on voice fundamental frequency responses to pitch feedback perturbations during sustained vocalizations
21
The Journal of the Acoustical Society of America. Neural representations of enhanced speech motor control in trained singers
22
PubMed Central. The neural control of singing
23
PubMed Central. Diagnostic Value of Acoustic Analysis and Inflammation Markers in Pediatric Vocal Fold Nodules