Essex University Develop New Algorithmic Model which will be developed by Phonak


Essex University have undertaken research which according to them reveals new information about how the impaired ear processes sounds. Phonak are taking their algorithms and will develop a hearing aid. It could lead to better hearing aids that deal with background noise:

They need special adaptions on their hearing aids to deal with all the jangly bling, hearing in convertibles and for the women to hear through their bleached, sprayed hair-dos. :smiley:

In all seriousness it’s interesting, but people like the ISVR in Southampton have been researching this kind of thing for years…

I thought the hearing aid industry had nailed background noise. But I know in my heart in can’t be true, because I still can’t hear in a crowd that well, although there has been an improvement.

Try defining that without using the word ‘noise’: (noise being subjective to the wearer) it’s quite a challenge.

‘Sound information which may or may not be speech that isn’t contextually relevant’

What provides the context?

Sure. Understand the point. But my point remains. Industry makes claims that hearing aids can cope in situations with large gatherings of people, but I don’t know if these claims are inflated. I know that from experience my own hearing aids are now good at attenuating noises from the side and back, but in a noisy room with lots of hubbub, the speakers voices in the immediate vicinity are not clear and audible above the general level of conversation. Marketing messages we get from the likes of Phonak, Oticon and co, is that this problem has been nailed. This to me, is the holy grail. I haven’t tried the Agil or Audeo S aids yet, so I don’t know if these would be any better.

I noticed when I was trialing the Rexton Cobalt that I seemed to hear better in noise than those with normal hearing. I would become somewhat the translator for the group.

If I had to grade them I would give the Rexton Cobalt (Costco) a 100 and the Resound Future a 95. The Futures seem to be improving, with no actual changes being made (I’m getting more used to them?) and right now I would say I’m almost on par with the normal hearing people, when in situations of high background noise.

It’s all about the programming of it and somehow we got the Cobalts just exactly right, and the speech in noise ability was amazing. The Futures are closer now and maybe we will get there but considering everything I can’t complain about them at all.

I noticed the Futures have some adjustments where we can bring up soft speech and lower background, more than one adjustment for both if I remember. So, we might try that for my last adjustment session coming up at the end of June.

Anyway, I think the speech in noise ability of these modern aids is very good but if they can improve on that then we will all benefit eventually.

The ONLY big step forward in this area I have seen in the last 7years (Post Oticon Adapto) is the Binaural Zoom function on the Ambra/Spice 9 aids. Everything else has been pretty much ticking with the edges: some of which people have liked, some of which my customers have described as ‘hearing via committee’…

There have been some big feature improvements (wireless, better feedback management etc), and some processing refinements (floating point linearity, speech enhancement) but actual changes to the signal to noise ratio ‘SNR’ as defined by Mead Killion have been few and far between.

Well in that case why aren’t the other companies scrambling to incorporate the Binaural Zoom function into their own hearing aids? I had high hopes for the “beam forming” technology but my audiologist was not that impressed with the feedback she’s gotten from clients who trialled the Spice 9.

I have a reverse slope hearing loss that starts at 60 db in the lows and climbs to 100 db around 2 kHz. I plan on trialling the Widex Clear Fusion HP once it becomes available in the US next month because I had a very good experience with my Widex Divas. Dunno whether or not to give the Spice 9 a shot. Any advice?

DUDE!!! After all you said to me about stereotypes :stuck_out_tongue:

Funny though. I only went to Essex once, and I can confirm the rumours about loose Essex girls are based on fact. :eek:

I’m not even going to read this article, because I am far too lazy.

But here’s the deal. Almost every manufacturer has developed their own unique fitting formula for hearing aids since the dawn of time, or at least dawn of hearing aids. There are a bunch of industry standard ones like NAL, BERGER, POGO etc and proprietary ones like e-STAT.

Ultimately a fitting formula is the starting point, it takes the test data and creates an initial ‘prescription’ for the fitting. Then the hearing professional gets to work crafting a successful fitting that the patient can both accept and derive useful (and measurable) benefit from.

I was kidding earlier about being too lazy to read the article. I did read it. Same old stuff that has been put out by dozens of hearing aid companies, and universities for decades. “We’ve cracked the code for amazing hearing, blah blah…”

Sorry, I’m not sure what is unique or clever about this. Just another researcher who thinks he or she has cracked the code. I’m not saying that their research is worthless, but it is just another bit of information in the steady evolution of hearing aids.

The research further admits that they have yet to gain clinical acceptance of their research. I also think it’s a bit smug to be talking about why people buy expensive hearing aids and don’t wear them merely because they created yet another in a long line of dozens of fitting formulas.

But hey, good luck to them, let’s hope their research does indeed move us forward, but I won’t be holding my breath just yet.

Ok ZCT, I love your cynicism, but at least Phonak have gone over there and are checking it out. That’s good enough for me.



Yeah they sure could use some better algorithms :wink:

How about:

Phonak have a patent trail on their method to prevent quick reverse engineering.

And it’s quite difficult to do: ask Oticon.

As for your Audiologist, it comes down to whether she’s really putting the Ambra/Spice 9 on the most motivated or highest spending clients. That will have a big bearing on outcome. That doesn’t sound like a reverse slope loss by the way, Soundrecover may be handy to pull back some speech for you. One other thing, try to get fitted with the Phonak rep present if your Audiologist isn’t keen. You’ll get the best set-up for a reasonable trial.

My point BTW related to the actual technical refinements that really ‘helped’ improve SNR, Beamforming is one of the few. Lots of the other features are just fluff from the manufacturers.

I said xenophobia: in any case, you’re having a conversation with a balanced Welshman: I have a chip on both shoulders.

Thanks for the tips. My mistake on calling it a reverse slope loss - I kept thinking it meant negative slope rather than reverse of ski slope.

I tried the Soundrecover feature when trialling the Phonak Exelias a couple years ago and no matter how much the audiologist tweaked the settings all the higher frequency sounds had an unnatural “shhh” sound which I didn’t like at all. My speech discrimination with both the Exelia and Naida hearing aids was terrible compared to my old Widex Divas. In fact my current Oticon Epoq Power RITEs don’t perform as well in noisy environments as my Divas did but the Streamer has made a huge difference for me.

I am hoping that the Widex Clear Fusion with the HP reciever will work for me with it’s combination of extra amplification in the highs, an effective background noise suppressing sound processor, and the M Dex.

I am all about improving SNR so I’ll talk to my Audi again about the Spice 9.


As I’ve said before, I’m an eighth Welsh. I trained to be an RHAD in Cardiff, my first trainer was Welsh, I vacationed with friends in Pembroke many a time, and I even dispensed aids in Wales for a year. Any jokes I made about the Welsh and their playful willing sheep are only in jest, I have nothing but nice things to say about the place and the people. My lovely grandmother sounded as Welsh as you could imagine.

Oh and I like Torchwood and Dr. Who, made by the Welsh. :slight_smile:

realize that using just three EQ points for low end control (750 and under), centering the EQ bands in very un-musical places, and using just 4 compression bands that group lows and mids together, does not work, then it’s all just band aids.

The info has been out there for decades. The HA makers just won’t change how they think.

Yeah, but they are not really thinking about music. They are thinking about communication with other humans. Music is secondary to the design of most aids. The vast majority of sounds in English occur in the higher frequencies above 1000Hz.

I am at the HLAA Conference in Washington DC and the North American CEO of Phonak announced this at the opening session last night. Lots of interesting studies also. One emerging study outcome is that hearing loss likely causes dementia…if untreated…in older populations.

Its a tough nut to crack this business of hearing in speech babble and other forms of noise.

As I was taught, noise is a complex of many different things,ie: repetitious stuff, non-repetitious stuff, speech babble all from various directions.

People with normal hearing use the ear shape (pina, concha) to provide direction response plus the brain acts like a filter providing directional, cadence, subject matter, frequency discrimination to tune in to the desired speech while rejecting unwanted stuff.

Hearing aids can only provide directional processing, some usually ineffective band pass filtering and filtering of repetious noise. They don’t come close to what the auditory cortex with its billion or so neurons can do.

I’m not sure science will ever get close to what the brain can do. Ed