An Open Letter to Hard of Hearing Musicians

Marshall Chasin
April 19, 2016

This letter is based on my 35 years of working with hard of hearing musicians at the Musicians’ Clinics of Canada.  Over the past few years an assortment of new technologies for hearing aids have become available that will allow hard of hearing audiophiles and musicians to reprogram their hearing aids once they leave the audiologist’s office. While on the surface this appears to be a good idea- one can fine tune the hearing aids in real life situations with real life musical instruments- there can be some drawbacks.

Recently the tools that audiologists use to program hearing aids have been made available for purchase by anyone. These include the fitting software for any number of hearing aid manufacturers, the interface box between the computer and the hearing aid called HI-PRO, and the necessary wires and cables that can connect the HI-PRO interface to the hearing aids.  These can be expensive, but given the cost of many hearing aids, these tools pale in comparison on a cost basis to the hearing aid expenditure.  Of course the cost is for much more than the physical hearing aid.  It consists of prescribing the correct amount of (possibly level dependent) gain and limits of the frequency response, follow-up, and most importantly the verification that the hearing aid actually does what it is supposed to do.

The short story is that I recommend that you leave the programming and verification in the hands of the audiologist. While this last statement sounds paternalistic, let me give you the longer story to explain why this is actually for your own protection.

There are many hearing aid parameters that need to be adjusted and verified by the audiologist. To be fair, most of these parameters are rather benign and the worst case scenario is for you to reset them to the initial settings.  These may include the frequency response, the various features of the compressors (including kneepoint thresholds, attack and release time, and compression ratio).  These parameters are probably well known to musicians and audiophiles.  Many use software for music composition and recording that has all of these parameters such as Logic Pro. I would suspect that the experienced musician probably knows as much about these parameters, especially for music, than many audiologists.

The one parameter that can cause additional permanent hearing loss (also found in the LogiPro and other music software) is the maximum sound that can be transduced through hearing aids. Audiologists refer to this as the OSPL90, or maximum Output Sound Pressure Level with 90 dB SPL input.  If the OSPL90 is set too high, then permanent future hearing loss can occur.

This is where the audiologist comes in, armed with Boyle’s Law.

Boyle’s Law is a 17th century discovery made at Imperial College in London, England.  Boyle found that in an enclosed volume the pressure was inversely related to the volume- the smaller the volume, the higher the pressure, and this includes the sound pressure as well.  We learn this in high school chemistry classes as part of the ideal gas law.  This is also why balloons tend to explode when one sits on them- the volume is decreased to a point where the pressure is so great that the balloon explodes.

The maximum output or OSPL90 that will be seen on the hearing aid fitting software is a number measured in a 2 cc coupler. That number is fairly accurate if your ear canal with a hearing aid in place has a volume of 2 cc but we rarely see that “average person”.  People with smaller volume ear canals such as children, have a much greater sound level generated at their eardrums than what the OSPL90 number will indicate.  This can easily be as much as a 10-15 dB difference, depending on the frequency.

Of the many tests that an audiologist uses to verify hearing aid function, there is a device called a probe tube microphone measurement. As the name implies this is a controlled miniature microphone situated at the end of a long silicon tube that can measure precisely what the hearing aid generates.  Without verifying the sound levels in a person’s ear one cannot be sure that the sound levels are too great.

In the near future, other technologies will be coming onto the market that again, will allow the hard of hearing audiophile or musician to program their own hearing aids. If you are so predisposed, go ahead and experiment with the frequency response, and compression characteristics, but also make an appointment with your local audiologist to have them ensure that the set-up does not generate sound levels that are too great.

Audiologists can verify that this is the case, and can even provide you with a frequency-by-frequency correction/calibration curve that translates the numbers on the computer screen to actual sound levels that can be generated in your ear canals.

  1. I agree with you totally. I’m a hard of hearing musician and an electrical engineer with a strong background in audio electronics. Because of what you wrote and because I do not have any knowledge of the physiology of human hearing, I would not dare to do my own programming of my hearing aids.

  2. Mr. Chasin makes a very valid point and that is overamplifcation. Aids have Limiters (MPOs) for a reason. Do not touch these, if you are self programming. In my experience, the MPOs are set properly. However, I have had two makes of aids that would overamplify, straight from the factory, so, self adjusting or not, if your aids are too loud, get that fixed right away.

    As you know all too well 🙂 I have been self programming for over a decade now, through maybe 5 or 6 sets of aids as my hearing deteriorated and hearing aid quality has improved. I made a lot of mistakes while learning, for sure, but I think that that there is a repeatable process for self programming. At least I found what works for me.

    Self programming is full of pitfalls, true, and it is far from simple. Yes, a person can self adjust, but it is not simply a matter of getting the interface and software and having at. Yes, if a person jumps in without knowing where the dangers lie, cause damage to their hearing. BUT, it is doable, if the person has the right gear, knowledge, patience, and follows a process carefully. No, you can’t play back some music on your computer to use as a test. It takes proper speakers that will make up to 104db, a proper room, real instruments and uncompressed recordings, plus 32 band EQs and a db meter (both cell phone apps, and beware there, too, if you have Android), and above all, a lot of critical listening skills and knowledge of audio terminology, to make your adjustments. And it will still not always sound the way you think, in the real world, just like in audio production..

    Over the years I have had other musicians contact me about self adjustment and, when they are faced with the realities of it, they change their minds. It is not easy, and it is tedious and often frustrating, especially is one tries to do too much in one sitting, especially without the proper equipment and knowledge.. Aids are not loudspeakers and aids software is not Protools. A person has to know the idiosyncrasies of hearing aids, their behavior in the ear canal, and the aids’ programming software, plus how all the sound processors work and their trade names. No, you can’t assume that what works in the recording studio will work for aids, and vice versa, but there is a lot of usable overlap that goes a long way to improving the sound of your aids. But I do not want to discourage someone from trying to self program. While a person may not be able to self program to “perfect” they can, if they pay attention, can greatly improve the sound of their aids, if they apply themselves to learning how.

    Yes, in situ programming has made great progress, yes, the science of audiology has grown by leaps and bounds, but, unfortunately, a lot of “sacred cows” impede the progress. And, sad to still have to say, there are too many audiologists who simply do not know how to run the software at a nuts and bolts level, beyond what they learn on line or in a three day seminar. For musicians, certainly, this is not enough, and this is why many of us look to self adjustment.

  3. I am a former musician and an electrical engineer by trade, in the pro audio field. I also do self programming. Quite often I just do simple things like change the amount of noise reduction, move programs around or delete ones that I find out in the real world are useless (like most of the factory “music” programs). I also do some tweaking of the levels to make them more comfortable. This usually means lowering the gain somewhere. It’s not for the timid though. Anyone with half an interest in it can locate the pieces of the puzzle to do it. I started tweaking with my wired programmer and as my aids went wireless, so did my tools (just like my audiologist — who is fantastic by the way).

  4. Although Dr Chasin’s warning is quite appropriate, MPO (formerly SSPL-90) levels are rarely hit, due to the wide dynamic range compression in all digital hearing aid circuits. What’s more, unlike with linear response typically found in analog aids (or if you program digital aids in true linear fashion), the MPO values are rarely hit, as the G80 values — the gain with 80dB SPL input — are often quite low.

    Separately, most software will provide the option to estimate the real ear response instead of using 2 cm³ Zwislocki coupler values.

    However, Dr Chasin’s admonition on deviation from 2 cm³ values is vitally important in diagnostic tympanometry when measuring acoustic reflex thresholds, as failure to compensate for canal volume can yield *false negative* results when screening for ANSD: All ipilateral reflex stimulation levels are calibrated to a 2 cm³ coupler; however the typical canal volume measured while performing tympanometry is typically around 0.5 cm³.

    Now, if the tympanometer thinks it’s putting 90 dBHL (the limit for ruling out ANSD per Berlin et al, 2010) into a 2.0 cm³ cavity but in fact it’s 0.5 cm³, the actual stimulation is 102 dBHL, which is a ANSD screening #FAIL. (Note that this admonition does not apply to the *few* tympanometers which use AGC to reduce the stimulus level after measuring the canal volume.). And Yes, I’ve seen at least two cases of missed pediatric ANSD diagnosis due to failure to compensate for canal volume.

    For more on this, see “Acoustic Middle Ear Reflexes: Simple, Underused, and Critically Important,” and note well the canal volume compensation values in Table 2.

    Dan Schwartz
    Editor, The Hearing Blog

  5. Hearing aid fitting signal processing software is very complex, in some cases much more so than pro audio dsp. I agree with Marshall that most folks, including musicians/audio engineers untrained in audiological fitting should not try to adjust the sound of their aids using current software interfaces.

Leave a Reply