Hacker News Reader

Comment by maroonblazer | Hacker News Reader

The article says the treatment focuses on increased speech intelligibility. Speech, depending on how you define it, can occupy a fairly narrow range of the audio spectrum that humans can perceive.

My hearing loss is mostly on the high end (10-12K and above) and impedes my ability to hear speech in noisy environments. More troubling has been the inability to enjoy music/sound in those upper ranges, resulting in the music lacking a certain sparkle and brilliance.

I hope this therapy regenerates hair cells across the entire spectrum.

View on HN

Check out Crescendo (http://refined-audiometrics.com/wordpress1/) -- an engineer/musician with a hearing loss similar to yours has developed a piece of software that's like a "super" hearing aid (non-linear compression per Bark band) that helps to bring the sparkle back for music.

Thanks for sharing that, its got some amazingly deep articles on audiology. It's really hard though, to find out what Crescendo actually is and how to buy it!

I think you need to email David McClain to buy a compiled version, but the source (not including the plugin wrappers, etc.) is here: https://github.com/dbmcclain/Crescendo-Hearing-Correction

I'm having a hard time understanding what this is still. Is it an equalizer?

It is a multi-band nonlinear compressor. It deals with each band of hearing (https://en.wikipedia.org/wiki/Bark_scale) individually. In each band, it elevates soft sounds to just past the threshold of audibility with someone's impaired hearing, to make it sound like the softest sound they can hear. And it elevates medium sounds a little bit less, to make it sound like a medium sound. And it elevates loud sounds not much at all. Hence the nonlinearity.

This reminds me of one aspect of my hearing loss that's been hard to explain.

Amplification doesn't fix everything. My hearing loss is complex, a mix of conduction problems in the middle ear and something neurological as well. I can hear very high and very low frequencies surprisingly well. There's a big bite out of the main speech band, though.

I find that speech, with background noise, that goes through a mediocre microphone, 200 - 3000 Hz pass filter, and then a digital compression codec, and then through a loudspeaker -- no matter how loud -- can sometimes be nearly unintelligible to me. It's loud enough. But I can't figure out which sound that's supposed to be. It's just too distorted.

I take my hearing aids out to listen to music, most of the time. Despite usually being able to better hear the melody and lyrics with the hearing aids in. Because I know the sparkle you speak of, and it's wrecked by all but the very best-quality electronic amplification. Some people think it's odd that I'm finnicky about quality speakers, etc. when I'm half deaf. But every decibel of SNR counts when I lose most of it in my head.

(And I too hope one day they can fix that for both of us.)

The other thing you miss with hearing aids, and even headphones, are the low frequencies picked up more by the body than the ears. Sunn O))), for example. I’d rather listen raw, too, but I raise the level somewhat, which can bother others.

What I probably need is a DAC for my sound system. The music streaming services like Apple don’t provide an adequate way to equalize the audio and balance to compensate for my hearing loss.

Edit: I keep hoping that Apple will do a better job here with their next release of AirPods Pro. One of the problems with Apple Music is they set the max volume too low. I can sort of understand the desire to help people protect their hearing, but their max level is less than IRL, and there is no way to bypass that setting.

Tidal is better. And there are plenty of YouTube channels I listen to at half volume, so it’s not the Apple hardware causing the problem.

I tend to listen to instrumental music. In part that's because I play an instrument, and in part it's likely due to hearing loss making words not that easy to understand. I also find I prefer vocals in foreign languages since there's no point in trying to understand them.

There's still plenty of good music out there.

I also find that my hearing aids aren't great for music due to how much they change timbre from boosting the high end, so I use regular ear buds instead.

As my own normal, age related roll off advances, I find all the good stuff is under 10 to 12Khz.

I mentioned different eq curves down thread. I don't have a lot of technical data at hand, but older radios from the vacuum tube era where they had response over 10 khz.

A good one, play that a respectable volume in a room, sounds pretty great!

What I'm kind of hinting at here is there maybe Equalization curves that can satisfy our need for that bright open sense that is missing when the higher frequencies are also missing.

That's not surprising, as what we think of as "presence" is usually in the ~4 KHz range, where our ears are the most sensitive. What's over 10 KHz may provide a sense of "air" but doesn't convey as much information.

Yup. Agreed.

Maybe the 4khz range is what was emphasized on older audio gear. I am thinking about the great overall impact that gear delivered is intriguing to me.

I think I've read something about the brain being able to use harmonics of the frequencies in question to "reconstruct" what's lacking. IIRC this is how Waves MaxBass works. This could be an added effect of your tube gear, which will be generating more harmonic content.

Maybe!

Another thought I had is how filters differ from an organic lack of response. Filtering wide band audio introduces artifacts. Those can be heard on AM every day in the US in the stations using brick wall type 5Khz filters.

A wideband signal run through something with low response is different!

And the cutoff matters.

No matter what the gear, a 5khz cutoff is crappy. 8khz can be good, and 10 can be hard for many to tell given good gear and quality source audio.

Once, I was able to listen to a great jazz recording on a older tube stereo. 50's era gear.

Playing vinyl on it was a terrible experience for both the listener and the vinyl. A recording done on tape was kind of amazing! The device was mone, pretty sure was a Zenith, and it had a cool baffle system on the speaker. One "tuned it for the room" by pulling a knob in and out. That shaped the low and mid range.

Was better than I would expect for that overall bandwidth.

Have you played around with an equalizer (hardware or software, doesn't matter) that offers rather extreme adjustments like +15dB or more, to see if you can approximate a flat source+ear system? Or is your loss more like a brickwall filter than a 12dB/octave slope for example?

I'm an audio engineer, not an audiologist, so I'm just shooting from the hip here as it's something I've been curious about in case I develop a similar issue as I age.

Another thing they could try would be a different eq curve overall.

I always find The Sound of Music as reproduced on older narrow band with am radios and phonographs interesting. A lot of that gear doesn't really produce anything over 10 khz.

But, where they emphasized frequencies in the spectrum is different from Modern equipment, which basically is shooting for flat response.

My grandfather's old radio, when played in a room have a comfortable volume, sounds really good. It's present. And even when I use the little transmitter to send it an idealized signal, it had a roll off starting around 8khz, the absolute high-end was maybe 12khz.

I never took the time to analyze it in detail, and kind of wish I had. The usual 62-100 hurts emphasis, like what you'll find on your typical loudness button on a 70's or 80's stereo was there. But it also shaped the sound in 3 to 8khz range.

The end result was a very natural sound the didn't overpower, felt like it should be there in the room with all the other sounds.

Wish I could say more but I'm not where I have any detail information. I can say the enjoyment of the music is different, and fulfilling. It's not bright, like those higher frequencies bring to the overall feel of the program being produced.

Robust, full, seem descriptive... clear is another one. Often, when I hear high bandwidth audio, through limited devices, it seems muddy, unclear, not precise. Those older radios are not like that. Well, the better ones anyway.

In any case, the eq is different, and that impacts our perception related to the overall feel of the sound.

Sorry for the typos. I'm doing this with the voice dictation. I may actually go back and revisit this myself. As I age up I'm getting be expected, natural roll off most all of us get in our later years.

I think one of the reasons for vinyl’s reputation is that it’s often played on old sound systems. Ones with real bass and treble controls!

Yeah I think that is true. However I was really speaking to something a little older than that, though vinyl could still be a source signal for it.

You go back far enough, and the gear starts to roll off about 10 khz, yet if you listen to a modern signal through it, it sounds good. That's what I'm getting at.

Please don't do this. This would work for a conductive hearing loss (ear drum or bones in your ear messed up, etc.) but not for a sensorineural hearing loss (like what you get from aging or from noise exposure). You need non-linear compression to compensate for loudness recruitment. At a basic level, when you have sensorineural hearing loss, soft sounds will sound way softer than they should (or be below the threshold of audibility), but loud sounds will sound about the same as they should. If you crank up the volume for the affected frequency bands to make it sound right for soft sounds, then with loud transients, it will sound way too loud, and potentially do more damage to your hearing.

Look at https://positive-feedback.com/Issue66/hearing.htm if you want to hack something together that will work better and not damage your hearing.

Quite right! I’m quite “deaf” in the sense that I can’t understand what people are saying - they sound muffled (and I can’t separate two people talking at once anymore).

But, I’m actually quite sensitive to loud noises!

What I need is equalization and compression, so I can hear the soft noises. I don’t need much, if any, overall gain.

Why can’t Apple provide that, they have all the necessary technology at hand?

Do hearing aids offer this sort of multiband compression? I've heard about hearing aids that act as Bluetooth headphones, so presumably your Apple device could send its unaltered audio over Bluetooth and then the hearing aid's custom programming could handle things from there?

Yes, they do (in fact, they have custom hardware to do it), but the power budget is very, very small, so there have to be tradeoffs made, both in terms of the processing and in terms of the actual transducer.

People are used to recharging AirPods frequently, and their hardware exceeds what is in a hearing aid. So, I wish Apple would enter this underserved market and stop just making "works with iPhone" deals with the hearing aid companies.

Thanks!

I have sudden deafness in my left ear (meaning: I just woke up deaf one day and they don't know why) in exactly the frequency range used for speech. Even more than a decade later it feels weird that I can still hear certain common background noises but humans are almost completely silent on my left side

> I have sudden deafness in my left ear (meaning: I just woke up deaf one day and they don't know why)

Ah, me too, but no sound at all. A specialist told me it was caused by a virus, but I never looked into which one, or asked for more info.. There was no treatment, except cochlear implant, which didn't sound too great. Hopefully it doesn't happen in the other ear also!

I miss being able to hear every little note with full range. A combination of construction noise and some loud music did it for me and I can’t even “imagine” music with full range in my head like I use to before my hearing damage and tinnitus.

Reply on news.ycombinator.com