An excursion through hearing aids

FIL suffered from severe hearing loss in his later years. Now he was a chap who had always been careful with his money, so it took a little effort to persuade him to throw some of it at a fancy hearing aid. And in my innocence, I had assumed that if you threw a bit of money at a fancy bit of computerisation and miniaturisation, you would get a result. In the event, there was improvement, but no miracle. I felt a bit bad about this, but to give him his due, he never complained.

So when ResearchGate thought I might be interested in some work (reference 1) done in Australia on tuning hearing aids for music, rather than for speech, which is the regular drill, I thought I would indeed take a look. Which turned out to include a quick swing through references 2 and 3 along the way. I should say that I have not read any of this material properly, but I have turned most of the pages.

Along the way I took an online hearing test – from the people a reference 8 – a gadget which gets you to play sounds of increasing volumes on your laptop for a range of frequencies – from which, if I can trust the calibration of this test on my laptop, I learned that my own hearing loss is in the range mild to moderate. While the lady from Boots a few years back settled for mild. And while reference 1 talks about a lot more men having hearing loss than ladies, say around half of us at age 50 and more or less all of us at age 100. Although he does not say how much loss there has to be to count at loss, as it were.

I also learned that most of this loss is of a fairly simple form. You just don’t hear quiet sounds. With the frequency spectrum sometimes being flat, sometimes not. Some people lose the low sounds, some the high - and some both, just leaving the central portion of the normal range more or less intact.

And a lot of hearing aids are rather simple too. They just amplify the inbound signal by some fixed amount, some fixed multiplier, with a cut-off at the top end so that you don’t get a pain in the ear. I think this is called linear. Noting that multiplying a small amount of loudness by a multiplier might still give a loudness below your threshold; stuff is still going to be missing. More cunning ones divide the loudness scale into a small number of segments and apply different amplification to the different segments. More cunning ones still do all this by a small number of frequency bands. And then there are dynamic considerations about how quickly the amplification responds to changes in input loudness. With all this stuff being managed by your audiologist - although I should add that, just to be on the safe side, the hearing aid will have a volume control, to give you back a bit of control. It may also have dial that you use to say what you are doing and what you want it to do; whether you are in the pub or in the concert hall.

Whereas I had just thought you could just map the input range of real sound to the output range of what the subject could hear, in some nice continuous and instantaneous way, without any trouble at all. Without being limited to simple multiplication. A thought which had quite overlooked the various distortions that are going to creep in – of, for example, the linear crescendos and diminuendos which are so important in western classical music. Or, if you vary things by frequency, of the relations between the various elements of a chord or the various over and under tones of a fundamental. What about the missing fundamental of reference 7? What about the distinctive timbre of a favourite violin or a of a favourite voice?

There is also the problem that most of us do not get around to seeing our audiologist until we have had quite a lot of hearing loss for quite a long time. Our brains have forgotten how to deal with all the background clutter, all the noise, and getting it all back again, all that background drivel from the next table in the pub, can be a problem. From which a whole new hare starts off: could we get the hearing aid to know what it was we were trying to listen to and to filter all the rubbish which we did not want to hear, out? So I am sitting in the Globe Theatre and it knows to cut out the sound of the helicopter hovering overhead. See, for example, reference 6.

I associate to reports of people who have been blind for a long time having trouble if their sight is restored, a miracle which is pulled off occasionally. But which is not always the miracle hoped for and I have read of at least one person who turned the miracle off: she had got used to things the way they were and did not like or could not get used to the change.

Then there is the complication that our perception of sound is very context sensitive, just in the way that our perception of colour is. It seems quite likely that hearing aids are disturbing that context, possibly in quite crude ways.

And as far as recorded music is concerned, there may well be interaction between what the audio engineers do to the raw input, perhaps in the way of sampling, clipping or filtering, to get what they give to you and what the hearing aid gets up to.

Unfortunately, I still have no idea how much hearing aid design (and result) is driven by the need for miniaturisation and whether one could do a significantly better job if, for example, one was allowed bulky equipment, wires and a few seconds delay to let the computers do their stuff, rather than having to try to do the job in real time, on the back of a postage stamp, on the hoof.

But all in all, an interesting diversion for a couple of days. Another tricky excursion into trying to get hold of the subjective experience.

With thanks to reference 1 for the abbreviations.

Reference 1: Hearing aids and music: experimental tests for better algorithm selection and design - Marcus Wigan, Peter Blamey and others – 2017.

Reference 2: Perceived Quality of Recorded Music Processed through Compression Hearing Aids - Naomi B.H. Croghan – 2013.

Reference 3: The compression handbook: Third Edition: An overview of the characteristics and applications of compression amplification – Banerjee – 2011. A handy introduction to hearing aid technology, in particular compression techniques.

Reference 4: http://www.starkey.com/about-starkey-hearing. The people who sponsored reference 3.

Reference 5: https://www.researchgate.net/.

Reference 6: http://psmv2.blogspot.co.uk/2014/08/antony-cleopatra.html.

Reference 7: http://psmv3.blogspot.co.uk/2017/01/virtual-pitch.html.

Reference 8: http://www.audiocheck.net/testtones_hearingtestaudiogram.php.