Effects of Filtering and Vocal Duration upon the Identification of Speakers, Aurally
Arthur J. Compton · The Journal of the Acoustical Society of America · 1963
The purposes of this study were to investigate (1) the effect of various conditions of high-pass and low-pass filtering of the voice upon the identification of speakers; (2) the effect of various durations of recorded segments of the voice upon the identification of speakers; and (3) the relationship of fundamental frequency to misidentification of speakers. Fifteen recorded segments of the vowel [i] for each of 9 speakers were presented to listeners. All listeners were familiar with the speakers' voices through daily contact. The 15 segments of the vowel differed only in duration. The samples of the speakers' voices were heard under 7 conditions of high-pass and low-pass filtering. The task of the listeners was to identify the speakers by writing the name of one of 9 speakers after the presentation of each vowel stimulus. The results of the investigation indicate that (1) durations of 1/40 of a second are sufficient for identifying speakers; (2) the greater the severity of filtering, the greater the duration of the sample of voice required for identification of speakers; (3) attenuation of frequencies of the voice below 1020 cps does not affect the ability of listeners to identify speakers; (4) attenuation of frequencies of the voice above 1020 cps substantially reduces the ability of listeners to identify speakers; and (5) there is an inverse relationship between the relative degree to which speakers' voices are confused and the range, in cps, between the fundamental frequencies of the speakers' voices.