- Some of the data of this study have been published in an earlier paper (P. C. Delattre, A. M. Liberman, and F. S. Cooper, “Voyelles synthétiques à deux formantes et voyelles cardinales,” Le Maître Phonétique, 96, 30–7 (1951). In that article we were concerned primarily to display the results of our syntheses graphically and to compare that graph with the IPA articulatory charts.
- The vowels which served as our models were produced by one of the authors, a phonetician whose native language is French. Naturally his conception of the color of the cardinal vowels might differ slightly from that of Daniel Jones whose pronunciation of these vowels on a well-known recording has long served as the standard. However, we should guess that the differences, if any, are very small, since for 11 of the 16 vowels the IPA offers French vowels as guides to pronunciation. (See The Principles of the International Phonetic Association, University College, London, 1949).
- Descriptions of the technique and of some of the results obtained with it are to be found in: F. S. Cooper, “Spectrum Analysis,” Journal of the Acoustical Society of America, 22.761–2 (1950); F. S. Cooper, A. M. Liberman, J. M. Borst, “The Interconversion of Audible and Visible Patterns as a Basis for Research in the Perception of Speech,” Proceedings of the National Academy of Sciences, 37.318–25 (1951); A. M. Liberman, P. Delattre, F. S. Cooper, “The Role of Selected Stimulus Variables in the Perception of the Unvoiced Stop Consonants,” American Journal of Psychology (In press); F. S. Cooper, P. Delattre, A. M. Liberman, J. M. Borst, L. J. Gerstman, “Some Experiments on the Perception of Synthetic Speech Sounds,” Journal of the Acoustical Society of America (In press).
- Summaries of and references to many of these studies are to be found in T. Chiba and M. Kajiyama, The Vowel: Its Nature and Structure, Tokyo-Kaiseikan, 1941. More recent studies are described or referred to in: H. K. Dunn, “The Calculations of Vowel Resonances, and an Electrical Vocal Tract,” Journal of the Acoustical Society of America, 22.740–53 (1950); Gordon E. Peterson and Harold L. Barney, “Control Methods Used in a Study of the Vowels,” Journal of the Acoustical Society of America, 24.175–84 (1952).
- M. Joos, Acoustic Phonetics, Language Monographs, No. 23, Baltimore, 1948, 83.
- A disk recording of the synthetic vowels described here is available. This recording includes the synthetic vowels that were selected finally as the closest approximations to the 16 cardinal vowels and separately the series of 235 experimental vowels from which the selection was made. The disk can be obtained, at cost, by writing to The Haskins Laboratories, 305 East 43rd Street, New York, N. Y.
- For comparison with the identifications of our synthetic sounds, we presented to the same subjects, in such a way as to duplicate the test conditions of Tests B and C, a group of vowels spoken by one of the authors. (We did not include the vowels of Test A because we doubted our ability to speak several of the vowels correctly.) In all of the identifications of these spoken vowels there was only one error.
- For the purposes of this part of the investigation, all the vowel formants, including those of /o/ and /u/, were made equally wide; when converted into sound, however, they were not of the same intensity, inasmuch as the playback is so constructed as to produce a reduction in intensity along the frequency scale amounting to 9 db per octave in the frequency range above 1500 cps. As we pointed out in Section 2, this attenuation corresponds roughly to the distribution of energy in normal human speech; it should be noted, however, that this rate of reduction in intensity is correct only on the average and hence is not necessarily perfectly appropriate for any particular speech sound.
- Fletcher pointed out in 1929 that early investigators had failed to detect the relatively weak second formant of the back vowels and had concluded, therefore, that these vowels were “singly resonant.” (See H. Fletcher, Speech and Hearing, New York, D. Van Nostrand, 1929). Our results suggest that these back vowels may, indeed, be single-formant vowels, but only in the very special sense that, because of “averaging,” formants 1 and 2 of the spoken vowel can be replaced, in any synthetic production, by a single formant which is located at an intermediate position.
- It is relevant here that /i/ is an extreme vowel (none has a higher second formant), and the kind of judgment-anchoring which so often occurs at the extremes of a stimulus series might account for the fact that a high-frequency formant was so often identified as /i/.
- Köhler matched single pure tones to the pitch of sung vowels and arrived at equivalent frequencies of 3480 for /i/, 2265 for /e/, 1140 for /a/, 470 for /o/, and 225 for /u/. These values are very close to those which we arrived at by the very different technique described above. Köhler's results are reported in E. G. Boring, Sensation and Perception in the History of Experimental Psychology, Appleton-Century-Crofts, New York, 1942, 373–4.
- For a specification of the frequency positions of the first two (lower) formants of French spoken vowels, see Pierre Delattre, “Un triangle acoustique des voyelles orales du français,” French Review, XXI, 6, May, 1948.
- Choose new content alerts to be informed about new research of interest to you
- Easy remote access to your institution's subscriptions on any device, from any location
- Save your searches and schedule alerts to send you new results
- Export your search results into a .csv file to support your research