Speech Distortions in a Hyperbaric Chamber

[Please read the article entitled 'The Helium Donald Duck Effect' before reading this one]

Divers undergoing recompression in a hyperbaric chamber sound funny. It has nothing to do with a DCS hit. It's caused by the reaction of the vocal tract to increased ambient air pressure. Studies of how high air pressures affect speech have shown that there is approximately a 4% loss in intelligibility for each atmosphere of pressure increase. The voice assumes a more 'nasal' quality.

Unlike the helium voice pitch effect, the hyperbaric chamber phenomenon is not caused by a change in the speed of sound. The speed of sound in an ideal gas is independent of its pressure. Rather, the phenomenon is the result of a change in the vibration properties of the walls of the vocal cavities and the soft parts of the throat when the ambient air increases in density (i.e. is compressed). For a given configuration of lips and tongue the vocal tract consists of a number of connected cavities. The peak energies of voice transmission occur at the various resonance frequencies of this complex system. The voice pitch is determined by the lowest such frequency, i.e. the fundamental, while the other qualities of the sound produced depend on the peaks at the higher frequencies, i.e. the formants. The first peak above the fundamental is the first formant, the next peak is the second formant, and so on.

When the air pressure in the chamber is increased, speech spectrograms show that the frequency of the fundamental does not change, but the frequencies of the first and second formants increase. Moreover, for most of the vowel sounds, the first formant shift is greater in magnitude than the second formant shift, which brings them closer together and causes the distortion of vowels. For example, as shown by the figure below (Fant & Sonesson, 1967), at a pressure of one atmosphere the spectra for the vowels 'e' and 'i' are quite different. However, at six atmospheres the shifts of the formants causes their spectra to become almost identical, thus making their sounds nearly indistinguishable. The frequency increases of the first and second formants explain the nasal quality of certain vowels at high ambient air pressures.

J. Adolfson and T. Berghage, 'Perception and Performance Under Water' (John Wiley and Sons, New York, 1974), Chapter VI.

G. Fant and B. Sonesson, 'Diver's Speech in Compressed Air Atmosphere', Mil. Med. 132:434 (1967)