Abstract
From a project concerning radiotelephone messages in aviation, a procedure was developed for the formulation of hypotheses respecting more intelligible and less intelligible speech sounds. From an analysis of words with poor ratings in articulation tests, several hypotheses are made, and as those same words appear subsequently in articulation tests of phrases and meaningful clusters of words, a study is made of the speech sounds undetected and of the mistaken responses of the listeners. Those missing or misinterpreted elements then are sorted and classified, and are related, when possible, to appropriate hypotheses concerning particular speech sounds. New hypotheses thus are formulated as a guide for selecting the vocabulary of a language of stock-phrases or clichés for specialized use.
It is suggested that the formulation can be further tested by means of specially designed articulation tests and possibly by means of a separate quantitative analysis combining intensity and duration of various speech sounds. Meanwhile the present formulation can be used in working toward immediate pragmatic improvement of recorded or stock messages; and operational tests can be conducted on the endresults achieved for various sets of fixed messages or for languages having specialized applications.
Get full access to this article
View all access options for this article.