An acoustical analysis of the cardinal vowels as spoken by Daniel Jones

An analysis is presented of the recording of the cardinal vowels made by Daniel Jones. A pitch synchronous formant tracking method is used to derive trajectories in formant space, which are then compared to the predictions of a scaling discretization of formant ratios (SDF). The results are used to measure the SDF's single parameter, yielding a value of 1.6/spl plusmn/0.03. A possible explanation is given for this value, in terms of desirable properties of vowels as discrete elements in an auditory communication system code.