A peculiar problem in communication arises when a talker is wearing an oxygen mask or a gas mask, in that both react on the vocal mechanism and distort the voice at its source. Thus, the loss of intelligibility when a gas mask is worn is much greater than would be predicted from the amount of muffling, and, even when the oxygen mask contains a microphone, the talker's enunciation is not, in general, clear. The reaction on the voice is in part mechanical, by constraining the facial muscles and obstructing the flow of the breath, and in part acoustical, by adding a new cavity onto the vocal system so that the talker has less control over the character of the sound. The requirements for good intelligibility during the use of such enclosures have been discussed in some detail in a previous paper (see reference 1).
The present paper concerns itself with an experimental analysis of the distortion produced by some small enclosures, and an evaluation of present theories of vowel production by comparing their predictions with the experimental results. Such a comparison is probably as severe a test of the theory of vowels as can be devised.
In general predictions from the present state of the theory of vowels give a useful qualitative picture of the nature of distortion by small enclosures, but are not in quantitative agreement with experiment. A criticism is given of the assumptions involved in the present theory, and some suggestions are made toward the future improvement of our knowledge of speech.
