4 ms·

>The frequency response of a healthy ear isn't flat across all audible frequencies, so you'd need to reference normal hearing... Isn't the normal frequency res

>The frequency response of a healthy ear isn't flat across all audible frequencies, so you'd need to reference normal hearing... Isn't the normal frequency response of a healthy ear dependent on the shape of the ear cartilage and the configuration of the ear canal and the ear drum? It would be different for every individual. Kinda like how Mom could always hear everything we did and said after bedtime while Dad, without even using his selective hearing, wouldn't even know we were still awake. >Are you talking about basically using active noise canceling to silence tinitus? Yes. Model the tinnitus and design the inverse filter based on the bandwidth and inject that inverse filter to become an active subtraction of the tinnitus response. I know it probably isn't possible because the noise is variable and originates in the brain instead of external to the ear so it is not easily quantifiable therefore the inverse operator will not be exact, optimum, or anything else. However, if you can model the signal then you should be able to design the inverse operator. Since the signal is just a band-limited input there is no reason why you can't dink around until you have a close enough model to be able to design the inverse filter which you would then inject as an external input thru an earpiece or some other sound generator. I'm a geophysicist with hearing problems, not an audiologist or otolaryngologist. It sounds reasonable to me. We deal with convolution/deconvolution and other signal processing as a regular part of the job process.

thfuran 3y ago
>Isn't the normal frequency response of a healthy ear dependent on the shape of the ear cartilage and the configuration of the ear canal and the ear drum? It would be different for every individual Yes, there's some individual variation, but human ears are all generally roughly the same structure, so there are known baselines for how they work. There's about 40 dB difference in minimum audible threshold between 50 Hz and 5 kHz. Same with near the top end of the hearing range, though where exactly that lies is more subject to individual variation (and age) >so it is not easily quantifiable therefore the inverse operator will not be exact, optimum, or anything else. However, if you can model the signal then you should be able to design the inverse operator. It's not originating from actual sound, so I think the approach fundamentally doesn't apply. Active noise canceling relies on destructive interference to actually physically remove the sound before it is perceived. Once you have the nerve signal, I think there simply isn't an anti-sound that would result in some other nerve signal that adds up to perceived silence.
- NotMichaelBay 3y ago
  I've definitely read anecdotes about people with tinnitus listening to noise to reduce it. It may not be the same effect as destructive interference but it seems like there's something at work there.
  - thfuran 3y ago
    As I understand it, for certain types of tinnitus, listening to a sound of the right frequency may temporarily suppress the tinnitus for some time afterwards. But the triggering sound is still audible, so it's not like noise canceling.

thfuran 3y ago

>Isn't the normal frequency response of a healthy ear dependent on the shape of the ear cartilage and the configuration of the ear canal and the ear drum? It would be different for every individual Yes, there's some individual variation, but human ears are all generally roughly the same structure, so there are known baselines for how they work. There's about 40 dB difference in minimum audible threshold between 50 Hz and 5 kHz. Same with near the top end of the hearing range, though where exactly that lies is more subject to individual variation (and age) >so it is not easily quantifiable therefore the inverse operator will not be exact, optimum, or anything else. However, if you can model the signal then you should be able to design the inverse operator. It's not originating from actual sound, so I think the approach fundamentally doesn't apply. Active noise canceling relies on destructive interference to actually physically remove the sound before it is perceived. Once you have the nerve signal, I think there simply isn't an anti-sound that would result in some other nerve signal that adds up to perceived silence.

NotMichaelBay 3y ago
I've definitely read anecdotes about people with tinnitus listening to noise to reduce it. It may not be the same effect as destructive interference but it seems like there's something at work there.
- thfuran 3y ago
  As I understand it, for certain types of tinnitus, listening to a sound of the right frequency may temporarily suppress the tinnitus for some time afterwards. But the triggering sound is still audible, so it's not like noise canceling.