Logo of the Physikalisch-Technische Bundesanstalt

New economic calibration concept for objective audiometry

25.05.2010

Up to now, for every short duration signal newly developed for audiometry the reference threshold of hearing had to be measured using a group of test persons. By integrating signals into classes, the time and effort for calibration can be reduced.

Short duration signals used in objective audiometry should evoke human physiological responses measurable as clearly as possible at small sound pressure levels and with short total measuring time.
To date, the "Normal hearing thresholds" for the calibration of audiometric devices are expressed by so-called "Peak-to-peak equivalent Reference Equivalent Threshold Sound Pressure Levels", depending on the span between minimum and maximum instantaneous sound pressure values. For both the PTB-conducted threshold measurements with test subject groups (A and B) and the clinical audiometer calibration (B) only minor instrumentation is required: a signal generator, an acoustic coupler or ear simulator including a calibrated microphone, a graphical display for the temporal waveform, and a standard sound level meter (see Fig. 1).

Determination of reference hearing threshold using a group of test subjects (A and B), and audiometer calibration (B)1

Figure 1: Determination of reference hearing threshold using a group of test subjects (A and B), and audiometer calibration (B)

One of the procedure's drawbacks is the lack - in terms of hearing physiology - of coherence with the hearing threshold. Even signals with very similar amplitude spectra yield widely spaced reference threshold sound pressure levels. Therefore, for a long time, researchers and development engineers were calling for a calibration scheme producing reference values with a physiologically adequate relation to the hearing threshold. A first step in that direction is to simplify the procedure by assessing equal reference values to signals having identical amplitude magnitude spectra, by integrating them into signal classes (Fig. 2). This is legitimate, since such signals yield identical hearing thresholds when presented by usual audiometric headphones, which was shown in a previous PTB study.

Classification of spectra

Figure 2: Classification of spectra

Contact person:

Johannes Hensel, Dept. 1.6, WG 1.61, E-Mail: johannes.hensel@ptb.de