Extended bandwidth hearing aids

There is some discussion within other recent threads on this forum about extended bandwidth hearing aids, so I thought I’d start a new thread that perhaps could stay primarily on this topic.

As a musician with mild-moderate hearing loss only at 6k & 8k, this topic interests me. But I also note that some other posters who aren’t musicians also ask why extended bandwidth seems not to be generally offered by many of the major hearing aid manufacturers.

Granted, some folks might not be able to benefit by an extended bandwidth due to their hearing loss or cochlear dead regions. But should that be the reason that extended bandwidth is not readily available to those who might be able to benefit by it?

I’m not talking about such features as “sound recover” or frequency-transposition that some hearing aid manufacturers offer, but rather real gain provided at frequencies from 10k through 16k.

I’ve read that Sebotek has a HD16 model hearing aid with a 14k bandwidth. Another company, General Hearing Instruments, uses the “Digi-k” circuit that claims to provide gain to 16k. I really don’t know much about Sebotek and General Hearing Instruments, but they seem (at least to me) to be smaller companies, and to my knowledge, no local audi or hearing instrument specialist fits these products in my geographical area. That’s the only thing keeping me from investigating these two options for my use, because I’m convinced that the success of a hearing aid is only partly due to the aid itself, with much depending on a good relationship with the patient and a hearing aid fitter who has extensive knowledge and experience with a certain brand/model of hearing aid and the fitting software.

It seems that most audiologists/hearing aid fitters only test up to 8k, and that a special audiometer is needed to test frequencies higher than that. So, a question for any audiologist/hearing aid fitter who fits Sebotek, General Hearing Instruments, or any other hearing aids with high bandwidth of which I am unaware: In order to be fit for one of these high-bandwidth hearing aids, does a person’s hearing need to be tested on one of these extended-frequency audiometers, or is it guesswork? (i.e., if someone’s hearing at 8k has a 50db loss, then does one just presume that the higher frequencies are worse, and then make an educated guess as to how much gain to apply at the high frequencies?)

I find this discussion of high-bandwidth hearing aids useful as well as perplexing. From my reading, it seems that high-bandwidth hearing aids have been available since the 1980’s or 1990’s, but it doesn’t appear that most of the larger hearing aid companies seem to have jumped on this band(width) wagon (bad pun!) despite studies showing the benefits of extended bandwidth not only for music appreciation but also for speech.

Any comments, thoughts, discussion would be appreciated! Thanks.

OK, How many octaves between 8 and 16khz?

What sort of power do you need to shake a speaker diaphragm at more than 10,000 times a second? The same diaphragm that has to deliver bass sounds.

What does this do for stress hardening at the welded copper drive rod?

What proportion of speech exists above 8Khz?

It’s called and engineering compromise. Technically it would be nice to do, practically though, it’s only really possible with back to back receiver and a crossover, which makes the whole thing twice as large, draw more power, twice as expensive and more likely to fail.

Um Bongo,

Thanks for your reply. By the way, I DO appreciate your expertise in this forum, as well as the other professionals who post here. Too bad some forum members give you guys a hard time.

I only know the answer to one of your questions. There is one octave between 8k and 16k. From the musical standpoint, that octave contains musical harmonics which help a person distinguish whether a particular musical pitch at the same intensity is played by, for example, a flute, oboe, or violin.

As far as speech energy in that octave, I’m not sure. I am aware that technically this octave is outside the “speech banana.” But some studies I’ve read do seem to say that octave is also helpful for speech comprehension, speech in noise, and sound localization. Perhaps those studies are wrong.

I don’t know anything about the technical aspects you mention, but it makes me wonder how Sebotek and General Hearing Instruments, for example, have supposedly achieved an extended bandwidth in hearing aids that are no larger than those of other manufacturers.

Yes, the primary reason folks seek hearing aids is for speech comprehension. But it would be great if manufacturers could also address musical fidelity as well.

Again, thanks for your reply, and looking forward to any other insights.

I agree completely with that last point, it would be great. However, fast acting chips designed to pick out very rapid changes in speech relative to background noise are practically designed to do the opposite. As are the transducer components to some extent.

Oticon advertises bandwidth to 10kHz, and Sebotek do their thing too, but what you won’t find is flat responses out to that level. Your Hi-fi will have a nominal response of 20Hz to 20KHz, with a decent AMP - stick on some tidy speakers with perhaps four driver units each and nice steep crossovers and you’ll get a great broad signal. Superb for orchestral and film scores, but a bit rubbish for speech. Especially speech that is mixed with noise/music by a 30 year old sound engineer who has never heard of Lemmy, Jimmy Page etc.

This is why we have so much trouble with modern Hi-Fi sources, as a casual rule of thumb, the more money somebody has spent on their Bang and Olufsen TV or their Kef Egg speakers, the better they think they will hear speech. However the flatter the response, the poorer the relative level of speech will be relative to noise. In terms of speech, every 1% worsening of the speech signal to noise ratio equates to approximately 10% worsening of speech comprehension. Which is why they end up listening to the TV or Radio in the kitchen that cost a tenth of the price, but gives them better speech clarity.

Now, if you accept the principal goal of hearing aid engineering is achieving speech clarity in background noise (because all the extensive studies on this tell the engineers that’s what people need most), the you find that the Hi-Fi argument runs possibly 2nd or even third behind making the aid work on the phone too.

Different manufacturers have isolated the music settings into both user selectable and automatic programs: to put the aids into a music ‘mode’ which basically shuts off the fancy electronics and speech recognition, turns down the gain and the feedback manager and makes everything a bit more linear. That’s (transducers aside) how good you can make it. Anything more will probably require a change in the speaker and mic technology: especially the conventional speakers - broadband mics are available now.

We’ve been slowly gaining the ability to amp out higher and higher, and we are dealing with several issues that Um bongo covered. We’re also dealing with signal processing. To digitally process a sine wave you need to process at least twice the rate - to process a 1 kHz sound you have to be able to sample at 2 kHz. We can do this. But to process a 16 kHz sound you are starting to talk about serious computing speed - 32,000 samples per second minimum. The power that takes can be better used for other things. As Um bongo mentioned, there are other things to focus on that have a heavier impact.

I’ve been playing and listening to music for more than 50 years and that’s a major contributor to my hearing problem. I got my first set of HAs in Feb of this year so I’m pretty new to this. I also had a concern about music fidelity, especially since I am wearing open fit HAs that don’t provide much bass. They have been adjusted with the bass boosted quite a bit and they are now acceptable for listening to the car radio and that sort of thing. I’m now hearing high frequencies that I haven’t heard for a long time so that’s good. Initially, I tried pumping music into them through a direct connect and bluetooth but I finally accepted that I was trying to get something out of them that they really weren’t designed to do. They help a lot with speech but they will never be a high quality music reproduction system. I rely on my headphones and speakers to produce the wider range for music and I can boost the lows (and highs) to whatever degree I want to provide a fuller sound (especially if I’m the only one listening). With the HAs on listening to music, I am hearing much more of the highs than I used to but they are probably cutting off around 8K (unless SoundRecover is doing something with the higher frequencies). With a good sound system and an equalizer, I could probably boost the top end so my own hearing would pick up a bit of what’s above 8K (as long as I was the only one listening). It might then be a matter of whether my ears and my brain are still capable of doing anything with those frequencies. Even if I had HAs with a much wider bandwidth, it might not make any difference. That may not be true for those whose hearing loss is relatively mild and recent but they aren’t the majority of hearing aid wearers. At some point the technology may be equal to the challenge but it still may only benefit a small portion of HA wearers.

I have found this topic interesting. I did quite a lot of work on high frequency (above 8kHz) hearing in my PhD in the early 80’s. I had to source and calibrate my own headphones to do this. I also used children under 10 years of age for my normative data, as high frequency acuity deteriorates young in life. I am sceptical that most musicians have used much in the way of auditory information above 8kHz since their teens or early 20’s, so the question becomes, what do they then do with the information.
I can’t comment on the brands mentioned, except to say that the bandwidth limiter on a hearing aid is usually in the microphone.

Erm, receiver.

Electret condenser mics have been around for quite a while that can comfortably do a couple of octaves above 8Khz with quite a flat response.

Getting a speaker unit to shake twice as fast at the same amplitude is harder though.

I notice in the new tiny CIC from Unitron, the speaker looks to be too flat to be of conventional construction. In conjunction with what we know* about the origins of Otolens, I’m guessing that they are now using panel excitation rather than conventional coil devices. This would be consistent with driving smaller volumes of air more quickly.

It opens up the possibility of having flat (or other shape) panel drive mounted on the rear of your CIC device. Or even the rear of the CIC moving in entirety to drive the sound. Possibly resulting in wipe-clean hearing aids without the need for wax traps.

*Otolens was derived from optical R+D where they were trying to change the shape of crystals through electrical charge and presumably alter the focal length of optics.