Frye Electronics, Inc.

by Robert Martin

Introduction:
Most hearing aid circuits that reduce background noise do so by cutting the low frequency gain and limiting the output. This paper suggests measurements you can make to see whether or not the hearing aid circuits are performing their job (increasing intelligibility) for the patient. These measures evaluate the speech information cues (real-ear gain) available to the patient for word understanding ability in quiet and noisy listening environments.

Summary
Give the patient the typical hearing tests. Then evaluate the patient’s ability to understand filtered speech. Select the hearing aids and make initial adjustments. Next, using a real ear system, generate an NAL target. Simulate a quiet listening environment by using a soft input level and do real ear tests to compare REIR (real ear insertion response) to the NAL target. Simulate a noisy environment by using a loud input level; again, compare the REIR to the NAL target. Inspect both sets of data for adequacy of amplification. Be sure the amplified sound is comfortable. Do sound box tests to evaluate the “cleanness” of the signal. Confirm this procedure by checking discrimination ability in a noisy listening environment.

Step-by-Step

1. Do a typical audiometric evaluation and find the patient’s maximum speech discrimination ability.
2. Since many “noise reduction” circuits cut the gain in the lower frequencies, evaluate the patient’s word-understanding ability for similar conditions. Use a master hearing aid unit (FONIX FA-10) and measure the patient’s speech discrimination ability through the -6, -12, -18, and HFE settings. Plot these scores. These filtered speech tests show how much the low frequency gain can be reduced without damaging word-understanding ability significantly.
3. Select, adjust, and fit the hearing aids.
4. Simulate a quiet listening environment by using a 50-dB composite noise input level. Overlay the REIR on the NAL target. Check for adequate amplification. The REIR values should match or exceed the NAL values.
5. Simulate a noisy listening environment by using an 80-dB composite noise input. If the hearing aid has a “noise reduction” switch or program key, activate it now. Next, run real ear tests and overlay the REIR on the NAL target. Check for adequate amplification in the 1000-6000 Hz zone, the zone most critical for word understanding. Adjust trim pots, digital code, etc. to best match the NAL target in the higher frequencies while reducing the gain in the frequencies below 1000 Hz.
6. Compare the response curve and RMS values obtained with the 80-dB input to the patient’s uncomfortable listening thresholds, UCLs. The response curve values give information at specific points; the RMS value indicates “total energy” in the signal. Some tolerance problems are observed in specific zones; other times the overall signal is just too loud.
7. It is important that the amplified sound have low harmonic and intermodulation distortion values. To evaluate the “cleanliness” of the amplified sound take the aid off of the patient and place it in the sound box, using an 80-dB input signal. The presence of intermodulation distortion can be observed by inspecting the response curve for smoothness. The amount of harmonic distortion observed at the use-gain setting can be quantified using a 70-dB puretone input signal and sweeping the spectrum.
8. Confirm the above procedure by measuring speech discrimination ability in a noisy listening environment.