Note: The featured image shows the picture of Georgia Brown, a Brazilian vocalist who is famous for her whistle register. You can listen to her whistle register here: https://youtu.be/Ia57VfDaESw?t=68

Voice registration is a phenomenon perceived as a quantal change in voice quality (timbre) superimposed on pitch, loudness, or phonemic descriptions. The spectral content changes from one register to an adjacent register, either due to source or filter characteristics. Here we address one of these registrations, namely whistle voice. How humans (and some large animals) can produce fundamental frequencies in the 2000 – 5000 Hz range is still somewhat of a mystery. One hypothesis is that the vocal folds can effectively be shortened by restricting or damping a portion of their length posteriorly (Kato et al., 2023; Fantini, 2024). This would be analogous to shortening a violin string with finger placement to increase pitch. While this hypothesis has not been disproven, it seems unlikely that a strong enough “pinning” force can be applied at controlled locations along the length with laryngeal adductory muscles. A second hypothesis is that a whistle tone may be produced with turbulent air in or near the larynx, as in a lip whistle (Edgerton, 2013). This hypothesis has also not been disproven, but a two-octave range of frequencies produced in the kHz range would require a highly adjustable acoustic resonator in the lower vocal tract. Such a resonator has not been identified and described acoustically.

Svec et al. (2008) and Echternach et al. (2024) showed that flow-induced vibration of the vocal folds is a viable mechanism up to nearly 2 kHz in female singers. The vocal ligament is likely to provide the tension in vocal fold tissues to reach these fundamental frequencies (Titze, Riede, and Mau, 2016). In a current study (Titze and Riede, 2025), a new hypothesis of whistle voice is proposed. A vibratory mechanism is proposed to explain how humans (and perhaps other species) can produce fundamental frequencies well into the kHz range while vocal fold length remains on the order of 1.0 cm. It is labeled as the epithelium-dominance hypothesis. Quoting Titze and Riede directly, “it follows a logical connection between depth of vibration and frequency in surface waves on solids and liquids. As depth of vibration shrinks at high frequencies, the tissue closest to the surface becomes dominant in stiffness”. This tissue is the epithelium.

Figure 1. Vocal fold tissue layer dominance for fundamental frequency control in chest register (M1) by muscle tissue, mixed (head) register (M2) by ligament tissue, and whistle register (M3) by epithelium tissue. After Titze and Riede (2025).

Mechanism labels M1, M2, and M3 were assigned by Roubeau et al. (2009) to distinguish vocal fold vibration patterns in humans. M1 refers to chest or modal register, while M2 refers to falsetto in males and head voice in females. The M3 mechanism has yet to be described mechanically, but perceptually it is known as the “whistle” register. Titze and Riede (2025) claim that M1 is characterized by thyroarytenoid (TA) muscle dominance, M2 by ligament dominance, and M3 by epithelium dominance (Figure 1). The concept of tissue layer dominance is important. Given that these three primary tissue layers of the vocal fold have different stress-strain relations, their normal modes of vibration will not be the same at a given vocal fold length. To obtain periodic vibration, one layer must dominate in self-sustained vibration. The other layers, if also in vibration, must entrain to the dominant layer. Figure 2 shows how tissue dominance changes with vocal fold strain (fractional elongation). Stress on the left panel and frequency on the right panel are expressed semi-logarithmically on the vertical axis due to their wide ranges, while vocal fold strain is expressed linearly along the horizontal axis. Strain is defined as the fractional elongation relative to the non-adducted vocal fold length. Note that the TA muscle dominates in mechanism M1, the ligament dominates in mechanism M2, and the epithelium dominates in mechanism M3. Stress in the ligament does not vanish at zero strain because zero strain refers to the abducted vocal fold length, which can be greater than the adducted length. Stress in the muscle is generally non-zero at zero strain due to internal fiber contraction.

Figure 2. (Left) Vocal fold tissue layer stress-strain curves plotted semi-logarithmically for three register mechanisms, and (Right) fundamental frequency for the same registrations. After Titze and Riede (2025).

Extremely high f_o seems to be easier to produce at younger age (Herzel and Reuter, 1997), and women are likely to produce “whistle voice” easier than men (Walker, 1988; Svec et al., 2008; Garnier et al., 2012). A hypothesis is that ligament and muscle properties are not yet fully developed prior to adulthood, suggesting that the system may favor the transfer of tension control to the epithelium. The shorter vocal fold length in females and children is also an obvious advantage for high f_o.

Extremely high f_o phonations have only recently been successfully modeled with computation because the required fiber stress in the frequency-dominant tissue is an order of magnitude higher than direct measurements have recorded (Perlman et al., 1984). In particular, the epithelium with a thin collagen fiber substrate in the basement membrane known as the lamina densa, may play a crucial role in offering high tensile stress (in the MPa range). No direct measurements of vocal fold epithelial stiffness are yet available. Titze and Riede (2025) argue that existing data on (a) the mechanical properties of other types of epithelial tissues and (2) species diversity in vocal fold epithelium morphology suggest that this thin outer layer could support extremely high f_o if it becomes dominant in self-sustained oscillation. A computer simulation with a sub-millimeter flexible membrane showed that fundamental frequencies between 2 – 3 kHz are possible with epithelial dominance. No vibration extended into the ligament or thyroarytenoid muscle tissue. The vibrational amplitudes were therefor extremely small.

REFERENCES

Edgerton, M E et al. (2013). Pitch profile of the glottal whistle (M4). Malays. J. Sci. 32, 78–85.

Fantini, M (2024). The Physiology of Vocal Damping: Historical Overview and Descriptive Insights in Professional Singers. Journal of Voice.

Garnier, M., Henrich, N., Crevier-Buchman, L., Vincent, C., Smith, J., & Wolfe, J. (2012). Glottal behavior in the high soprano range and the transition to the whistle register. The Journal of the Acoustical Society of America, 131(1), 951-962.

Herzel, H., & Reuter, R. (1997). Whistle register and biphonation in a child’s voice. Folia phoniatrica et logopaedica, 49(5), 216-224.

Kato H, Lee Y, Wakamiya K, Nakagawa T, Kaburagi T. (2023). Vocal fold vibration of the whistle register observed by high-speed digital imaging. Journal of Voice. Oct 6.

Perlman, A. L., Titze, I. R., & Cooper, D. S. (1984). Elasticity of canine vocal fold tissue. Journal of Speech, Language, and Hearing Research, 27(2), 212-219.

Roubeau, B., Henrich, N., & Castellengo, M. (2009). Laryngeal vibratory mechanisms: the notion of vocal register revisited. Journal of voice, 23(4), 425-438.

Švec, J. G., Sundberg, J., & Hertegård, S. (2008). Three registers in an untrained female singer analyzed by videokymography, strobolaryngoscopy and sound spectrography. The Journal of the Acoustical Society of America, 123(1), 347-353.

Titze, I. R., & Riede, T. (2025). A proposal for epithelial dominance in extremely high fundamental frequency vocalizations. The Journal of the Acoustical Society of America, 158(2), 1283–1295. https://doi.org/10.1121/10.0038968

Titze, I., Riede, T., & Mau, T. (2016). Predicting achievable fundamental frequency ranges in vocalization across species. PLoS computational biology, 12(6), e1004907.

Walker, J. S. (1988). An investigation of the whistle register in the female voice. Journal of Voice, 2(2), 140-150.

HOW TO CITE