. Carnegie Institution of Washington publication. EXAMPLES OF VOWEL ANALYSIS. 147 which ought to have been considered by investigators many decades ago— led to a hunt for a method of introducing it into the analysis. Only a method in which one factor of friction is used was developed; the method of approximating the value of this factor is, moreover, rather crude. As shown in Chapter VIII, provision should be made for at least two factors, one to represent the suddenness (p. 121) of the puff from the glottis and the other to represent the average friction in the vocal cavities. Even in its pre
. Carnegie Institution of Washington publication. EXAMPLES OF VOWEL ANALYSIS. 147 which ought to have been considered by investigators many decades ago— led to a hunt for a method of introducing it into the analysis. Only a method in which one factor of friction is used was developed; the method of approximating the value of this factor is, moreover, rather crude. As shown in Chapter VIII, provision should be made for at least two factors, one to represent the suddenness (p. 121) of the puff from the glottis and the other to represent the average friction in the vocal cavities. Even in its present condition, however, the method gives fair approximations to the truth instead of the absolutely false results given by the other methods. We will now analyze an actual vowel curve; the task is a more diffi- cult one because the real composition is unknown, and there is no test of the accuracy of the results. Figure 129 gives a piece of curve from the middle of the vowel [o] in "fourteenth" of the Mitchell vowel record. Fio. 129.—Waves from [o] of ";. Fig. 130.—Group of waves from figure 129. In the first place it is evident that the cur\^e consists of a series of waves (or wave-groups, p. 40), each of which comprises three vibrations of differ- ent heights. The object of the analysis is to find how such a group was produced. We might begin the analysis at any point—as most investi- gators have done—and include one group of three vibrations. Such a procedure would be justified by the overtone theory of the vowels, accord- ing to which it would be indifferent where the beginning of a wave is placed. Quite aside from the manifest incorrectness of this theory, it is sufficient to find by trial that the results of the analysis of speech curves (not musical curves) differ radically with any change in the point of beginning. We must therefore determine how the vibrations are arranged within a group, whether a large one is followed
Size: 2960px × 844px
Photo credit: © Book Worm / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: ., bookauthorcarnegie, bookcentury1900, bookdecade1900, bookyear1906
