| Abstract: | ![]()
The purpose of this study was to investigate the relation between the traditional concept of formant frequencies and the controversial concept of optimal octaves in the perception of the vowels/i, a, u/. The vowels (produced by a male speaker of General American English) were filtered through eight bandwidths (80-160, 160-315, 315-630, 630-1,250, 1,250-2,500, 2,500-5,000, 5,000-10,000, and 10,000-20,000 Hz) and presented to two groups of subjects. One group performed similarity ratings on pairs of filtered and nonfiltered stimuli; the other identified the individual filtered and nonfiltered vowels. The optimal bandwidths derived from the data of the two groups were compared to the formant frequencies of our speaker, to those reported in the literature, and to the optimal octaves published earlier by other authors. The results showed that there were specific bandwidths which allowed for correct perception and identification of each vowel, and that these bandwidths were compatible with both the optimal octaves and the formant frequencies reported in the literature for each vowel. |
