Bone-Conduction Speech Testing
Randall C. Beattie, David W. Gager · Ear and Hearing · 1980
Our major purpose was to obtain threshold force levels for spondee words and corresponding voltages using the Radioear 6-70 and 6-72 bone vibrators. A total of six bone vibrators were used, three of each type. Threshold force levels in dB re 1 dyne were measured on a Bruel 8 Kjaer model 4930 artificial mastoid. Two spondee thresholds, one with each type of vibrator, were obtained on each of 48 normal-hearing listeners. Because of the irregular frequency responses of bone vibrators, it is not clear which calibration signal is most appropriate for bone-conducted speech. Therefore, in addition to the 1000-Hz calibration tone preceding the Auditec spondees, a speech noise and broadband white noise were dubbed on tape at the same level as the tone. The results showed that both vibrators gave very similar mean threshold force levels (35.9 d6) for the 1000-Hz calibration signal. When the broad-band and speech noises were used as reference signals, the 6-72 required about 5 d6 greater intensity than did the 8-70 vibrator. The concomitant voltage measurements indicated that markedly different voltages are required to drive the 6-70 and 6-72 vibrators to threshold force levels. Similarly, quite different voltages were observed among the three calibration signals. Inasmuch as the clinical value of the voltage calibration method has yet to be established for bone-conducted speech, the use of these voltages should be used with reservation. Transfer functions were obtained with input signals of 1000-Hz and broad-band white noise, as well as total harmonic distortion values for a 1000-Hz tone, and these data suggest that bone-conducted spondee thresholds may be conducted up to 70 d6 without exceeding the linearity limits for either vibrator. It is concluded that additional research is required before a preferred vibrator or calibration signal can be recommended.