Simulating Acoustic Properties of Ear Canals

The acoustic properties of the ear canal can be characterized by measuring the real ear unaided response (REUR) [6 (link)]. This response is measured with a probe microphone inserted into the external auditory canal and shows the sound pressure level at the eardrum after the presentation of a well-defined broadband sound stimulus. Differences between individual REURs therefore represent differences in acoustic properties of individual ear canals. For instance, the acoustic effect of an ear canal with a radical cavity can be simulated in a normal ear canal by filtering the incoming sound stimulus using the difference of the REUR of a normal ear and the REUR of a cavity ear. The filtered sound stimuli, presented to a normal ear, should result in the same distribution of sound pressure at the eardrum as in the original radical cavity [6 (link)].
We used Dutch speech recordings (two male and two female speaker sentences based on the VU98 sentence material [9 (link)], filtered to simulate the acoustic properties of six ear canal conditions: two normal ear canals, two pre-operative conditions (ear canal with exostosis and radical cavity), and two post-operative conditions (canalplasty and revision radical cavity surgery with reconstruction of the posterior ear canal wall). The REURs of these six conditions were measured using the REM module of the Affinity 2.0 Hearing Aid Analyzer platform (Interacoustics, Denmark). Figure 1 shows the REUR results of the six conditions, presented as a real ear unaided gain (REUG, being the difference between the incoming broadband stimulus and the REUR). Six filters c.q. simulated conditions were built based on the differences between these six REUGs and the average REUG of a normal adult ear canal (see Table 4.6 in Dillon H (8), page 110 [10 ]). The seventh ‘reference’ condition consisted of the unfiltered speech material. We included sound samples, using English sentences but the same filters, comparable to those who were presented to the participants in the sound files.Fig. 1

Measured real ear unaided gain (REUG) of all conditions: two ‘normal’ ear canals (N1 and N2), a pre- and postoperative condition of a patient with exostosis who underwent canalplasty (Ex Pre and Ex Post) and a pre- and postoperative condition of a patient with a radical cavity who underwent a revision surgery with cartilage reconstruction of the posterior ear canal (C Pre and C post) (dark lines). In each window the average adult REUG is also depicted (Dillon) (light line). The REUG data are depicted on the same scale from 100 to 7000 Hz on the frequency x axis, and −25 to 25 dB (gain) on the y axis