If this uploads correctly, for all who may be interested, here is a pretty thorough fitting session I am testing. As the title suggests, it was done entirely by AI. I knows the history of my hearing and of the ear conditions and surgeries that have been done. I tried to give it any info I could think of that may sway its decision making. If anyone wants, I can send or upload the entire chat log, it has a lot of the details and the “reasoning” from it thinking through the decision making process.
So how do you feel making the audiologist redundant
I think it’s a great idea, but more importantly does/will it actually work and be any better then what you could do yourself using the algorithms built into the manufacturers software? like your testing right now, what are your preliminary observations?
Would surely be nice to see all the thoughts and explanation of AI. So please place here or DM to read the whole story of discussion. Never to old to learn and see if AI already is trying to become the best kid in school.
At the other hand as @tenkan says can you give any comments later on, if you sorted out if it was helpful for you more then using the probably also more or less “AI method” from the Phonak specialists/ engineers.
And I still assume that our beloved workers in the audiology- field as tenkan do not have to be afraid of “ So how do you feel making the audiologist redundant”.
I have only been using the program for maybe 12 hours. It seems good so far, but I will put myself in different environments over the coming days to test further. But I will come back every couple of days to give an update.
Also, I will post the entire thing here as well. Be warned, it is incredibly long. Does anyone now if there is a way I can post it that wouldn’t make it tae up so much space, unless someone wanted to expand it? Or is that done automatically?
Here is the best way I think. It is just an attachment PDF containing almost all of it. You can see I made it start fresh because it had some confusion when I had upload my audiogram via picture, so I went back and corrected everything by uploading full resolution screenshots instead.
I believe it’s safe to say that an audiologist’s role would be challenging to replace by AI. You’re in good hands. However, I do think it will become a much more valuable resource in the future. The thing about hearing is that it’s entirely personal perception. Comparing the results of countless others can provide a quick and generally accurate assessment. The personalization aspect will always be lacking. I may be completely wrong, and no one will have a job in 10 years. Who knows, lol.
@Cahje How did you phrase your question to GROK? I am assuming that the fitting interactive guide was created on GROK3 or am I way off base here. I have entered a lot of trial and error requests and have received pages and pages of information but I have not been able to recreate your interactive program.
Update: It appears that I am at least started in a positive direction just by telling GROK that I want to create an interactive program to fit my……hearing aids using the Target software. I am sure that it will be a couple of steps forward and one back each each try. This is my first try at using AI for anything other than a simple search.
I don’t know whether to thank you or not for getting me started on this
And thanks to the forum member who messaged me who was doing the same thing on chatGPT.
I think you have to very careful with the information you enter (or fail to enter) Interesting that both Grok3 and chatGPT recommended that I use NAL-NL2 and avoid Adaptive Phonak Digital 3 unless the other prescriptions have been found to be unacceptable .
But where is AI getting it’s information from for ones fitting, from audiologists/manufacturers or random bits from people’s observation on what they thought was best for them?
Apologies for the delayed update—it’s been quite a busy time for me. Let me share my initial thoughts. Using Grok3’s advice with the programs, I can confidently say it significantly improved performance compared to my limited skills in adjusting frequency gains.
However, one major issue I noticed was Grok’s recommendation to turn the wind block either completely off or nearly off in most programs. While this isn’t a problem across the board, it’s particularly unbearable in spherical mode. I plan to adjust this setting once I return from my current trip, which should be tomorrow evening.
Aside from that, the sound quality has been excellent, and I’m genuinely pleased with the results so far.
@raylock1 Grok began exhibiting the behavior you mentioned once I informed him that I wasn’t relying on him to provide feedback for my audiologist. I explained that I had access to the software and would be utilizing its input to conduct testing on program fittings that were developed using its logic and reasoning.
In my opinion, the primary reason why Grok defaults to not opting for the proprietary fitting formula is that NAL and other well-established formulas have been extensively tested and proven effective. These formulas have yielded tangible results that can be used to draw conclusions. Conversely, Adaptive Phonak Digital lacks publicly available data to gain insights into the specific goals or targets of the formula.
In my specific case, Grok arrived at a conclusion based on the probable cause of my hearing loss. If my memory serves me correctly, the formula selected for my left ear was a formula that had demonstrated superior results when combined with an individual who had hearing loss resulting from surgical procedures. Initially, that formula was not Grok’s recommended approach. However, after I provided a detailed explanation of my hearing loss history, Grok determined that the original approach was not the most suitable solution as it had anticipated.
