Originally published at: https://boingboing.net/2019/09/12/digital-hearing-aids-are-produ.html
…
So, a middle ground exists that xould be worth exploring: digital control of analog systems.
That means that the signal for the user to lusten to is run through completely analog lines whose parameters are controlled by digital circuits.
Collin Cunningham did a cool proof of concept with an Arduino controlled distortion pedal back in the day that demonstrates a similar concept.
I don’t know anything about hearing aids but Adam Savage some weeks ago posted a video recommending a particular brand saying it had very good acoustic range that can be changed on the fly, so i assumption is that the lousy audio quality might have more to do with lower end/cheaper aids being sold?
They also have little competition but are expected to be sold on a market, leading to less than stellar results.
Either we need to systemically incentivize much more R&D into providing better hardware, or open up the market to a lot more choice. There’s no reason a device as simple as a hearing aid should cost the same as a phone, but they often do.
or just better quality digital processing. There is so much that can be done with DSP on a chip that fits inside your ear that would require an entire rack of analog components (if it could be done at all). And digital technology is entirely capable of producing the full spectrum of sound that the ear can perceive - if it is designed well. Sounds like the hearing aid makers are taking shortcuts.
Yeah, digital isn’t the problem at all. Lazy development and medicaid profiteering with few incentives to provide a good product is a much more significant part of the problem.
Eh…? This makes no sense. Even the oldest MP3s I can recall still used 44 kHz sampling rates, just like “an old CD player”. And of course “an old CD player” has exactly the same sampling rate as a “new CD player”, these things being standardized. And despite what Neil Young might say to the contrary, 44 kHz is entirely adequate for almost everyone. Heck, back in the day people thought 128 kbps MP3 was practically CD quality. If you were stuck dealing with uncompressed WAV for some reason and had to deal with 22 kHz, you noticed - that’s telephone quality, that is.
Maybe there are other things wrong with digital hearing aids, but as written this smacks of the ol’ audiophile woo.
Agreed. The Slate writer clearly doesn’t know shit about audio technology. Blaming the low quality on the hearing aids being digital, rather than on the digital specs not being up to snuff, is ridiculous.
It’s actually quite odd that he was chosen to write this article—his bio says that he “works in the banking industry. He has a Ph.D. in political science from the University of Toronto and is currently writing a book on nationalism.”
That’s not to say political scientist bankers can’t know anything about audio tech (though this one doesn’t), but there are certainly folks more qualified to author this article. There are, contrary to popular belief, engineers capable of top-notch writing.
they cost FAR MORE than a phone. I had to get one 6 months ago.
That snagged my brain too, but then I realized he was likely conflating bitrate with sampling frequency, which I can’t fault any layperson for doing. Once I swapped those details and continued reading, it continued making sense. (Apart from the whole “digital sounds inherently less good, it’s just Science™” thing, that I kinda also can’t really fault the modern layperson for believing.)
But at that point, yeah, at least run the piece past someone who can tell hear the difference between 128k and 256k without checking the file size.
As someone who actually HAS hearing aids I can say that a lot of improvement needs to be done. They help but need to work better, especially in environments with background noise (restaurants) or which have lots of sound reflection (ie, bare walls, hallways, etc). Yes, I have cell phone control but it could be better.
I’ve also used hand-held radios (expensive commercial ones) in noisy environments on a daily basis for decades and can say through experience that the analog ones were much more clear on transmission clarity than are the digital ones.
More competition is coming in the H/A market, but I want improvement now.
Homebrew hearing aid using Russian vacuum tubes on Hackaday in 3… 2… 1…
WWII generation elderly still go to audiologists where they are paying $5-7k for h/a.
Costco carries 3 major brands and is far less expensive. They also have the best
price on h/a batteries.
without wishing to be overly facetious - it’s a serious matter - these listeners may have to learn to vote with their ears. But how do they do that? The audio industry is largely voluntary, consumerwise, whereas the hearing industry’s consumers pretty much have to put up with what’s on offer. The push-pull is in the other direction.
Also - all sound is analog on receipt. Until we become digital.
I’m surprised that people who are partially deaf can pick up the subtle nuances lost when you encode down to a moderate bitrate MP3. Even people with normal hearing struggle to catch the differences much of the time.
Woo, indeed. Digital hearing aids do a LOT with very little (space, power). Analog processors were dumber, worked worse. And, get this, the users are NEARLY DEAF.
One caveat though, is that Costco’s HAs are not the latest version of a said companies technology. That’s how they are able to sell them for a lower price. Just something to be aware of. I found this out after the fact.
I tend to go on rants about psychoacoustics. Less frequently now that I’m not currently in the training delivery business, but still.
I liken it to crappy paperback books printed from beat up plates made from fifth generation negatives with the press running as fast as possible on the cheapest paper they can get away with. Grab one of those and any decent hardcover. Look at the same lowercase letter in each book. Can you tell what letter that is? No doubt in your mind? So what’s the difference, then? How could one be “harder to read” or somehow “more fatiguing” than the other?
So thinking of it that way, it’s not that big a leap to imagine that when the signal from the sensor is degraded, the brain (and various brainlike bits along the way) needs to rely more on various subtleties that our “lossless” algorithms have declared superfluous.
Telephone quality audio is typically worse: 8 kHz with 8-bit logarithmic PCM samples, unless you’re dealing with cell phones or VoIP equipment that supports HD Audio (22 kHz is typical there).
For others’ information, more competition is coming in the hearing aid market because while hearing aids are currently classed as medical devices- and have a lot of regulatory hurdles associated with being a medical device- the FDA is now required to create and regulate a class of over the counter hearing aids. Supposedly, that is what will encourage other companies to enter the market.
So soon people will not need to go to a doctor to get (and fit, and set up) a hearing aid.