Like many of my colleagues, I have worked with musicians since the old days of linear Class A hearing aids- what some would call “crummy peak clippers”. Reading through many of my old blogs it won’t take time to realize that one thing that was good about the 1980s was the lack of analog-to-digital converters in hearing aids, which is the weak point when it comes to modern hearing aids and music. But at least back then we didn’t really need to worry about setting the compression for music listening.
Many of the rudimentary compression strategies in the 1980s and early 1990s were designed to prevent a hearing aid from entering compression prematurely. We had adaptive compression, frequency-dependent compression, and even high level compression. With today’s compression strategies it may be slightly more daunting, but not significantly so.
My gut feeling (and I think that I am right on this) is that the compression for a music program should be set similarly to how the compression would be fit for a speech program. Compression has more to do with compensating for the pathology of the sensori-neural hearing loss rather than the nature of the input stimulus, so it shouldn’t really matter whether the input is speech or music.
If the input is very intense (such as with many forms of music), then the hearing aid would be functioning in a different region of the non-linear input/output curve than if the input was quieter, such as for music.
I receive phone calls and emails almost every week from colleagues and consumers seeking advice on how to set a hearing aid for music. Many have spent hours “tweaking” the settings on the compressor of the hearing aid, to no avail. On the few occasions where this tweaking resulted in an improvement, these same new settings should be applied to the speech program as well, since the person would now have a better overall fitting. Again, the bottom line is that the compression for both speech and music are similar.
Surveying many of the hearing aid manufacturers will result in some rather bizarre suggestions for fitting music. Many specify frequency response changes and others suggest compression changes. In the vast majority of cases, there is no science behind these suggestions.
Some manufacturers suggest a wider frequency response for a music program than for a speech program; others suggest an enhanced mid-frequency and others suggest an enhanced low-frequency end. These suggestions are silly and for the most part should be ignored. If a wider frequency response is at all possible in a hearing aid this should be equally applied to both speech and music programs.
In the case of compression, it is no less silly. Some manufacturers suggest a more linear response while others tend to recommend that we compress the heck out of music. Again, I can see no rationale for altering the compression characteristics specifically for a music program versus a speech program.
The only exception is that I can see a different music program for live music versus recorded or radio based music. Recorded media such as CDs and MP3s, as well as almost all “jazz” radio stations, compress the music as much as they can. Many people, including me, don’t notice this, but many other audiophiles, musicians, and those with sensori-neural hearing losses, tend to notice this a great deal. In these cases, having a linear (or even slightly expanded) non-linear type of processing may be quite useful
In contrast, for live music, the settings should be similar to speech.
The (almost) bottom line-“a music program is a waste of a program”. The settings should be identical to that of speech in quiet and any advertising about the usefulness of a music program is merely advertising.
Now, here’s the real bottom line…. If the manufacturer has taken steps to ensure that the analog-to-digital converter has not distorted the music (e.g., Widex Dream; Bernafon Live Music Plus; Unitron -6 dB/oct microphone), then the compression and frequency response for speech should be the same as the compression and frequency response for music.