Inspiratory or reverse phonation refers to the production of voice when air is inhaled from the mouth and nostrils into the lungs. This way of phonation occurs naturally during laughter, sighs, and crying [1]. It has also been used to achieve special vocal effects by singers, shamans, and ventriloquists, and is common in the vocalizations of other mammals and birds. Furthermore, it is a useful treatment exercise in voice therapy for several voice disorders [2]. However, despite its relevance, few physical studies of inspiratory phonation have been conducted, and its underlying physics still needs further clarification.

In general terms, inspiratory phonation follows the same mechanism as expiratory phonation; namely, a flow-induced oscillation of the vocal folds that generates an acoustic wave. Nevertheless, it also presents some differences compared to the expiratory case. For example, experiments on human subjects have reported an inversion of the oscillatory pattern of the vocal folds, lower closed quotient of the glottis, higher open quotient, larger oscillation amplitude, and larger noise components in the acoustic signal [1, 2]. Other differences have been reported in vocal tract morphology [3] and articulatory timing control [4].

This note focuses on the phonation threshold pressure (PTP) in the inspiratory versus expiratory cases. In expiratory phonation, PTP is defined as the minimum subglottal pressure required to initiate the vocal fold oscillation [5]. This is a critical parameter as it is a measure of the “ease of phonation” –the lower the threshold, the easier it is to produce vocal fold oscillation. In the inspiratory case, the subglottal pressure is negative, but we consider its absolute value in order to avoid dealing with positive/negative signs. Our recent experiments on mechanical replicas of the vocal folds have shown a larger PTP in the inspiratory case, around twice the expiratory value [6]. Such a relationship appears consistently across different configurations of the experimental setup, and we argue that it may be a consequence of the asymmetric geometry of the glottal channel in the expiratory vs. the inspiratory directions.

It is well known that the profile of the glottal cross-sectional area is smoother when entering the glottis from the trachea than at the abrupt aperture to the epiglottal region [7]. Therefore, an airstream flowing from the epiglottis down to the larynx should suffer more pressure losses due to turbulence than when entering the larynx from the subglottal region. These higher losses would require a higher value of the glottal air pressure to initiate vocal fold oscillation, which could explain the experimental results.

Naturally, additional experimental data are required to explore the effect of other laryngeal parameters related to tissue biomechanics and aerodynamic factors. Nevertheless, the results so far suggest that phonation is more challenging with an inspiratory airflow, which may provide a potential physical explanation for why we typically speak during expiratory airflow. In other words, the intrinsic dynamics and asymmetry of the larynx make expiratory phonation the path of least aerodynamic resistance. This bias toward expiratory phonation demonstrates how certain broad aspects of voice production may be guided by self-organizing physical factors, prior to any neural control.

(This research has been supported by CNPq, Brazil) 

References

[1] Vanhecke, F., Lebacq, J., Moerman, M., Manfredi, C., Raes, G.-W., & DeJonckere, P. H (2016). Physiology and acoustics of inspiratory phonation, Journal of Voice 30, 769.e9–769.e18. https://doi.org/10.1016/j.jvoice.2015.11.001

[2] Orlikoff, R.F., Baken, R.J, & Kraus D.H. (1997). Acoustic and physiologic characteristics of inspiratory phonation, Journal of the Acoustical Society of America 102, 1838-1845. https://doi.org/10.1121/1.420090

[3] Moerman, M., Vanhecke, F., Van Assche, L., & Vercruysse, J. (2018). Vocal tract morphology in inhaling singing: Characteristics during vowel production—a case study in a professional singer, Journal of Voice 32, 643.e17–643.e23. https://doi.org/10.1016/j.jvoice.2017.08.001 

[4] Ng, M. L., Chen, Y., Wong, S., & Xue, S. (2011). Interarticulator timing control during inspiratory phonation, Journal of Voice 25, 319–325. https://doi.org/10.1016/j.jvoice.2010.01.001 

[5] Titze, I.R., Schmidt, S.S., &Titze, M.R. (1995). Phonation threshold pressure in a physical model of the vocal fold mucosa, Journal of the Acoustical Society of America 97, 3080–3084. http://doi.org/10.1121/1.411870

[6] Lucero, J.C. (2023). Physical modeling of inspiratory phonation, NCVS 2023 Conference on Self-Organization in Voice and Speech (online). 

[7] Scherer, R.C., Titze, I.R., & Curtis, J.F. (1983). Pressure‐flow relationships in two models of the larynx having rectangular glottal shapes, Journal of the Acoustical Society of America 73, 668-676. https://doi-org.ez54.periodicos.capes.gov.br/10.1121/1.388959