Nice job guy! I would love to hear the Lab Recordings when you get them up.
1 Like
Ah I forgot to upload the audio files. This is done now 
1 Like
One of the best “speech in noise” scores and one of the worst “own voice” scores. Pick your poison type of device I guess. Probably a nice fit if you’re a mute introvert.
With firmware 1.0.1?
Interesting to see how many best/worsy pairings there are:
Phonak Lumity = Best Speech in Quiet/Worst Own Voice
Starkey Evolv = Best Speech in Noise/2ndWorst Own Voice
Oticon Intent = Best Music Streamimg/2nd Worst Own Voice
ReSound Nexia = 2nd Best Own Voice/Worst Speech in Quiet
Oticon Real = 2nd Best Speech in Quiet/2nd Worst Music Streaming
Pick your poison indeed!
Interesting to see that the Widex Moment, which draws raves from users for the quality of its music sound is only middle of the pack for Music Streaming. I guess that’s different than live music?
Also interesting to see how many get better Music Streaming on Initial Fit than Tuned Fit.
Also, let’s be honest, who cares about the speech in quiet at this point, it’s a low hanging fruit. That’s not what most have issues with.
Well, this just shows how individual we all are and that it’s hard to rely on lab testing. I am currently trialling Intents and while I find the speech in noise and speech in quiet both very good, the music streaming is really bad for me. I don’t have my audiograms loaded yet but I have sloping mid/high frequency loss (moderate to severe IIRC). The music quality is so bass heavy and lacking in mids that I am applying a -12dB bass cut and a +3dB mids boost on the streaming EQ to get anywhere near an acceptable sound. I find the MyMusic program unusable and if I go with Intents I will have to create a custom program (other users have recommended using the DSL v5 prescription as a starting point).
So IMO this is clear evidence there’s no substitute for personal trialling.
At least Widex SmartRic seems to be well rounded after tuning.
I’m suspicious of some of the results. For example, Own Voice scores of Signia Styletto [which I own] show almost no difference, before and after tuning.
The thing is they have a special setup for that, where during [proper] initial fit it learns your voice, will treat and process it differently, and the improvement is significant.
1 Like
I disagree, Reginald, it’s a big deal for me. In fact, I think that it has a big effect on overall sound quality.
Just my 2 cents.
1 Like
Spot on. 
I’ve been through this fitting and it’s exactly as you say.
Sure, I get that, I should be more verbose.
What I meant is that technically it should be easiest thing to do (although looking at Nexias it seems that sometimes even the best can fork it up).
Also, I’m really angry at Widex for not going with BLA/Auracast for SmartRics.
1 Like
Also, I don’t understand the scoring system. How average of 4.3/4.8/2/2.6/2.8 gives the result of 4.6 (Lexie B2). (it’s 3.3 if the rules of common math still apply)
Same is true for the Signia Pure C&G. No difference at all for Own Voice between Initial Fit and Tuned Fit.
I guess it’s hard to learn the voice of a mannequin. 
Unless Steve has something different to tell us, I think that we have to consider the Own Voice Tuned Fit scores to be invalid for the Signia products.
1 Like
Must be a weighted average.
I would hope that it’s documented somewhere as it supposed to be a ‘scientific’ scoring.
Yes, it is documented in their white paper that it is a weighted average. Each of the Speech scores (Quiet & Noise) is weighted at almost 3 times the amount of each of the other 3, or each Speech score being almost equivalent in weight to the other 3 combined.
1 Like
Do you have link at hand?
Sorry. I meant to include it. Here it is:
1 Like
Here’s a message from Andy Sabin, HearAdvisor’s Scientific Advisor:
For our “own voice” metric, we are making a measurement of Real Ear Occluded Insertion Gain (REOIG) — the difference at the eardrum between the open ear and the ear with device in but powered off. This measurement has a good correspondence to user reports of subjective occlusion. More in our whitepaper.
That said, it only captures one aspect of what can make self-voice annoying. There is (1) the occlusion effect (body-transmitted-self-voice sound getting “trapped” in a sealed ear canal) and the (2) amplified effect (self voice amplified by the hearing aid mic). Our metric only attempts to quantify (1). The device you mention claims to reduce (2) and our metric would not detect that. It is a valid limitation of our metric to point out. We were willing to accept that limitation because we believe that (1) is a much larger detractor than (2), and we are not aware of any reliable way to measure (2).
We’d welcome any feedback on how to better communicate that.
1 Like
Nice work.
Unless my eyesight is going the route of my hearing, I’m having trouble finding a “Log Off” option at the top of the Forum, etc pages, so am still suffering with being able to get back to what seems to be a single Log Off option for the entire site. Can that be added to the top of each section?