Compensating for one's normal age-related high-frequency hearing loss...

A

Ally-007

Member
Our ability to hear higher frequencies (above about 1kHz) tails off increasingly (typically) as we pass the age of about 20. One graph I've seen suggests that the average 70-yr-old man's abilty to hear 8kHz frequencies is reduced by about 60dB! So music mixed and mastered by an 'old' person, without some form of compensation, is likely to sound way too bright, to a 20-yr-old listener. Can anyone offer any suggestions for us older producers, re mixing and mastering, with this in mind? I guess compansating with EQ is probably the obvious approach, but what do you think?
1711635578554.png
 
Last edited:
I'm 55 and don't really have an issue. I have substantial hearing loss from decades of live performances and just age. I think once you are trained and have experience mixing, your brain just compensates to some degree. If things are recorded properly, you just know you don't need to pull up the high end 12db so you can hear it I suppose. LOL
 
Well... my eyes AND ears are going lol.
So, I've plugged both my computers into 56" in my living room.
I rely on movie DVDs, streaming, youtube, etc... for a sound reference. But, I only adjust the sound at the amp, not the computer!

Here's my studio :)IMG20240328153314.jpg
 
I think the important thing is to realise that the top note of a piano is about 4K, so the harmonics of that note cover at least the 8K, so our usual EQ frequencies are well within the range. I'm now 65, and it starts to tail off at 9K, and is gone totally by 12. Looking at the mixes, I see not much at all above 13K, and that's only evident with cymbal heavy music and synths. I don't think my mixes are bright because I've not stuck in much up there in the first place. I could have accidentally done it on things but nobody has ever mentioned it?
 
I don't think that the graph is all that accurate. I'm over 70, and the last time I checked I could still hear some 10K. My left ear is better than my right. The problem is more the tinnitus that sometimes crops up more severely due to sinus issues. There's no way that I'm down 60dB at 8k on good day, tho. I know some people who probably are... they are using hearing aids. I haven't reached that level yet. I'm watching TV right now, and the level was bouncing between 54 and 60dB, so not exceptionally high.

That said, from 10 to 20K is one octave. It's the shimmer of cymbals, bells and triangles that I miss, not a lot of the main music frequencies. If anything, I tend to roll off the top 5kHz, not accentuate it, even if I can't hear it. Most condenser mics have a boost in the 10k-15k region so if anything, I think that probably evens things out.
 
I've made some progress, and the main stumbling block was a filter that could slice off at a certain point and not gently roll off like a normal variable Q filter. I didn't find anything in Cubase but I did in Audition - an FFT filter with lots of cut at an exact frequency and above. I've created a clip from an existing pice that has lots of top end from cymbals and harmonics from an organ sound - it was tricky, because I can't actually hear what I can see, but I checked on my phone app and it could hear what I can't.

The results will be interesting I think. Because most filters on screen are displayed logarithmically, the important area is squashed, so in audition, I displayed it as linear. I looped the section of the track, and at each loop, changed the cutoff. I started really low, and was surprised how while 'dull' - even a splash cymbal was still a splash, then I went up in 2K increments right up to 20K. I could hear a difference between cutoff at 12K and 14K, but above that, I could see the extra segments coming to life but not detect anthing. I'm really interested to hear the comments from younger less limited folk. I'm already realising that once your age (or damage) prevents you hearing differences it's very tricky to imagine what is missing. I figured I could easily 'guess' the missing stuff and estimate how much extra top end to add, EQ wise, by watching the EQ in the display as i turn it up, but I'm not certain I can - I think maybe EQ'ing what I can hear, and not see might be best. I really don't know. Hopefully I will get it edited tomorrow and then you can all listen. I wonder if it might also reveal if YouTube are limiting Frequency response in their compression algorithm? Need younger ears with opinions. I'm quite excited by this, as I think we've all been guilty of assuming hearing loss from age or injury is easy to deal with and easy to correct?
 
When you're a youngster, you don't appreciate how good your eyes and ears are.
I like to think that I'm doing ok for age related deterioration, but in truth I'm probably average.
My grandma always complained that TV presenters don't speak up properly.
Will we ever realise that our faculties are fading away?
 
Slouching Raymond said:
Will we ever realise that our faculties are fading away?
Click to expand...
I realize it every time I hit a golf ball and it disappears into the sky. That's why I have neon green, yellow and orange golf balls in my bag. I keep waiting for the eye doctor to say "time to get those cataracts taken care of". 6 years in and every time I go to the eye doctor, she says that they really haven't progressed since your last checkup. And I go DAMN! She just laughs at me.
 
Ive never met anyone who hasn't been amazed. Trouble is our NHS wont do them till one gets very bad, then people are stunned, but have to wait for the other one for too long. You constantly hear them saying wow, just wow.
 
Video done. Ironically, the monitors in the the edit suite didn't reveal the hum on my audio, discovered afterwards - but the test tone sections are mercifully free from the low level buzz.

Just shows how 5" monitors really don't do that well low down.

I'm quite happy with my hearing apart from an unexpected dip around 10K - which I didn't know about. Try it and see what you think?
 
Even more worrying for different reasons, I just got home and tried it on my big screen. I know it's got no bass (no soundbar) but when the test tones started 10KHz was silent, but so were all the ones above it. It has no response at all above 10K - I guess as my hearing tails off I've never noticed! On the music I can't hear any change above 8K!
 
Age 63, I took a test and my hearing tails off pretty much over 8k. I also have tinnitus which is probably a symptom of the high frequency hearing loss. The challenge comes when mixing, there is a tendency to level up the higher frequencies. I end up using some visual aids (Voxengo Span, Waves PAZ), I know you should not mix with your eyes but, what other options? I guess most my listeners are similar age and can't hear the difference one way or the other.
 
rob aylestone said:
I'm going to find some music, then band limit it at varying frequencies and pop it on youtube so we can compare our hearing - us ancients vs the young ones and hear the differences
Click to expand...
Hey Rob, I'm so glad that somone was interested enough to investigate this subject by putting together an actual test. I just watched your video. Very nicely done! I'm knocking on 70 and listening to your test, at fist, I thought I could only hear up to 6k, but then realised I could hear up to 12k (just), if I cranked up the volume after the 6k tone. But 12k was the absolute limit for me. It was very revealing - thank you! I would like to find a test that produces a graph showing one's level of hearing ability at all frequencies from say, 1k to 22k. I think my approach when mixing, from now on, will be to avoid boosting anything above above 6k (maybe lower). It will be interesting to hear what people of different ages say in the video comments. Hopefully they won't be shy about stating their age.
 
Last edited:
When you have an NHS hearing test, they only concern themselves with speech frequencies - so they don't even test the high stuff. There are some on-line hearing tests that use the NHS frequencies, that play tones randomly that you can do, but they're not calibrated in any way as everyone has different headphones.
 
Ally-007 said:
if I cranked up the volume
Click to expand...
That's the only bone I have with Rob's test. Everyone sets their volume levels differently.

rob aylestone said:
When you have an NHS hearing test, they only concern themselves with speech frequencies - so they don't even test the high stuff. There are some on-line hearing tests that use the NHS frequencies, that play tones randomly that you can do, but they're not calibrated in any way as everyone has different headphones.
Click to expand...
I used to carry out hearing tests as a young technician, in an isolation booth, both ears separately, up to 20KHz.
The subjects wore headphones, and the volume level was fixed. At rising frequencies, a warble would get gradually louder. As soon as they hear it, the subject presses a button.
Then it gradually got quieter. When the subject could no longer hear it, they released the button.
 
After searching (within my time constraints), I found a few free online hearing tests, today, but they all gave vague results, such as: "You have some hearing loss." It would be great to find one that gives you an informative results graph to download. Surely, there must be one, out there, somwhere!
 
I can barely hear 12k on my system... but that could just be my old speakers?
This isn't exactly a controlled environment lol ;)
 
I think the snag is that if we accept the usual 'rule' that we can't hear less than a 2dB change, then our monitors or headphones would skew the results. co-incident dips and peaks would cancel. A friend of a friend is an audiologist, and he tells me the machines used are very clever in how they pick the sequence and levels. The operator can tell where people are unsure and adjust up or down, so the next time that tone appears it's XdB different, so they can roughly find a baseline for a particular loss frequency, then refine it. If you had some sealed headphones, you could with patience draw a very precise curve. Then, if you could get a measurement mic into the ear cup, you could plot the headphone response, and then adjust you curve up or down to cancel out the headphone's one? I suspect we could do this, but ordinary people couldn't.
 
You must log in or register to reply here.
Back
Top