Skip to content
HearingTracker
Log in

An update about Phonak EON vs Fortell from the HearAdvisor Lab

6 replies389 viewsLast reply

Link to this topic
Reply
  • With gain and occlusion matched, the difference in scores between the two devices shrinks substantially. When matched, our ease-of-understanding predictor (the HearAdvisor model) is tied (5.6 vs 5.6), and the objective intelligibility predictor (HASPIv2) gives Fortell a slight edge (6.1 vs 6.3).
  • Most of Fortell's published lead in noise comes from two things other than the : a more favorable first-fit gain profile, and a more occluding default dome.
5 people in this conversation3 min readJump to latest

Would it be to costly to do tests on wider variety of hearing loss? High frequency, cookie bite, reverse slope. I watch most reviews on hearing aids. Overall basic review 1vs1 or individuals personnel thoughts. I assume most people would more interested in a review or test based on a closer representation of their hearing loss. But there is also a lot of people individual likes and or preferences. In general thanks for all your information you provide.

Edited

In reply to @AbramBaileyAuD

<A bit shocked>

I am Phonak's and Roger user for over 10 years, but I didn't expect that difference would be so small.

I thought Fortell would still have a significant advantage over Phonak Audéo Eon Sphere 90 even when both would be measured with the same earpieces.

I wonder if Audéo Eons will get firmware upgrade later 😅

Edited

In reply to @SlightlyMuted

@SlightlyMuted said:

Would it be to costly to do tests on wider variety of hearing loss? High frequency, cookie bite, reverse slope.

But that means that the tests will be carried out for a very long time, because dozens of combinations are created, each of which takes a lot of time. And there are other devices to test...

Not to mention re-tests and 1:1 direct comparisons.

I only wonder if HearAdvisor will include testing (and does it make sense).

Edited

In reply to @SlightlyMuted

@SlightlyMuted said:

Would it be to costly to do tests on wider variety of hearing loss?

Yes. And we are behind on testing all the products for just one hearing loss. Each product is probably 30-40 hours of work including all the pre organizing, lab time/recording hours, and data reporting. We are applying for a large govt grant though, and one of our goals is testing for more hearing loss types.

Could the lab run this same closed fitting test on a non-Sphere Eon for a closed fitting baseline without ?

In reply to @AbramBaileyAuD

Cochlear dead-spot research is obviously very difficult because of the issues with obtaining ‘a look’ at the pathology while you’re currently testing the hearing.

The ISVR used to model the potential loss shapes by reverse engineering a ‘basilar membrane’ to develop the same dead spots. In terms of hearing aids, you have a KEMAR plus couplers and reference mics. You also have a zonal sound field generator within an anechoic chamber and you have control of the input noise/speech signal.

Surely the rest of the process can be pretty effectively automated by telling the AI to resolve the speech and noise per channel/zone then measuring results of each device in respect of active/directional denoising. Once the AI has captured enough data points, wouldn’t it be able to extrapolate a facsimile/transfer function of the test hearing aid in various loss settings/venting/canal volume/impedance guises for each aid?

I appreciate the aids are more complicated than they used to be, but the actual real time modifications they apply to the sound in each channel aren’t that revolutionary, simply because the combined output would sound alien if the phase slipped or other artefacts were introduced.

I’m sure you’re all over this, relative to my spitballing above. The other aspect of this which might be a consideration, is that the human factor means that you can derive the absolute optimum performance, but people don’t always choose that. Like the Vinyl/CD debate: even though CD is demonstrably superior, people still spend a fortune to listen to Vinyl because they ‘like’ the signal corruption that it brings: though I like the idea that you’re trying to produce a datum for hearing aid performance.

Related discussions