The -6 dB/octave microphone for music:
This post is the latest in a series about the ways that the hearing aid industry has responded to improve the quality of music for hard-of-hearing people. And like the other methods, this one has the desirable side effect of improving the quality of the hard-of-hearing person’s own voice.
The previous entries in this series included a definition of the problem (part 1), which can also be found here; four clinical strategies to improve an existing hearing aid for music (part 2); and something based on the definition of “dynamic range”-i.e., it’s a range and not a static number (part 3).
This week’s post is about an approach, which to date only Unitron Industries has had the foresight to implement, that uses a microphone that is less sensitive to the lower-frequency sounds (of music and of speech).
This is a relatively low-tech innovation that any manufacturer can use, and, as we will see, it has no downside.
Modern hearing aids have a broadband microphone- that is, the microphone has almost the same sensitivity for the low-frequency sounds as for higher-frequency sounds. As a rule of thumb, this helps ensure that the microphone noise (internal noise floor) is kept as low as possible.
Consider a common hard-of-hearing client who has relatively good low-frequency hearing thresholds, but poorer high-frequency hearing. This is a common configuration for people with presbycusis, noise-induced hearing loss, and music-induced hearing loss. These clients may require little or no low-frequency amplification, and in many cases can be fitted with a non-occluding fitting where the lower frequencies actually enter the unoccluded ear canal, bypassing the hearing aid. In my clinical practice, the majority of my clients fall into this group.
If these people are fitted with a non-occluding hearing aid, which allows the intense, lower-frequency components of music to bypass the hearing aid and enter the ear canal directly, then we may think that we have an optimal hearing aid fitting. After all, the lower-frequency intense sound energy of the music can no longer overdrive the front end A/D converter of the hearing aid.
The problem is that the low-frequency, intense sound energy of music is also going through the hearing aid route. It is true that the lower-frequency components of music will not make it through to the client’s ear; the low-frequency sound energy will go backwards out through the non-occluding ear canal and be lost to the person. But it is also true that the higher-frequency distortion products of this intense low-frequency energy will stay in the amplification chain and be heard by the hard-of-hearing person.
We need a method to reduce this possibility, which is where the -6 dB/octave microphone comes in.
There is a commercially available hearing aid microphone that is designed to be 6 dB less sensitive at 500 Hz, and 12 dB less sensitive at 250 Hz, as compared with a flat or broadband hearing aid microphone. At 1000 Hz and above, this microphone behaves identically to a broadband hearing aid microphone. This is sometimes referred to as a low-cut microphone.
Instead of using a broadband microphone, the hearing aid designer just pops one of these less sensitive microphones into the hearing aid. In this scenario, the intense, low-frequency components of music have been turned down before anything gets to the A/D converter at the front end of the hearing aid. Since many people don’t require much gain at 500 Hz (or 250 Hz), this is typically a difference that makes no difference.
But (and I love “buts”), the internal noise level of the hearing aid will increase significantly in the lower frequencies. This is one reason that some manufacturers are hesitant to incorporate this low-tech innovation.
And here comes another “but”…. But, modern hearing aids have expansion, and increasing the action of the expansion circuitry reduces this low-frequency internal noise to a level that is identical to that of a broadband microphone.
Below are three curves. The first shows the internal noise level spectrum for a broadband microphone (red), the internal noise level spectrum for the -6 dB/octave microphone (purple or magenta color), and the internal noise spectrum for the -6 dB/octave microphone with expansion implemented (black). Note that with expansion, there is no deleterious effect on this less sensitive microphone. The red and black curves are virtually identical. (Special thanks goes to Mark Schmidt of Unitron Industries for performing the internal noise spectrum measurements.)