Skip to content
HearingTracker
Log in

Oticon Neural Noise Suppression

This topic is closed.

70 replies6.7k viewsLast reply

Jump
Link to this topic

OK, thanks @flashb1024 !

By the way, just to round off the earlier discussion of when the NNS Difficult default is 6 and when it’s 8 dB, it looks like for my situation (my hearing loss and personalization settings), Genie 2 sets my NNS Difficult default to 6 dB in the default VAC+ P1 program, but if I add a Speech in Noise program, then Genie 2 sets the NNS Difficult default to 8 dB for Speech in Noise.

Below is the screenshot of P2 in my setup which is Speech in Noise. The differences between that and my regular P1 default program is that it defaults the Difficult configuration to include Moderate (my P1 defaults Difficult to exclude Moderate situation), and it ups the Easy NNS to 4 dB (from 0 in P1), and Difficult NNS to 8 dB (from 6 in P1).

The other difference I notice between my P1 and the Speech in Noise program is the gain curves in the Fine Tuning section. I think Genie 2 boosts up my gain curve a little bit more for Speech in Noise.

image

@Volusiano said:

Difficult default to 6 dB in the default VAC+ P1 program, but if I add a Speech in Noise program, then Genie 2 sets the NNS Difficult default to 8 dB for Speech in Noise.

So, NNS is adjusted per listening conditions, and the extra reserve of the More 1 is beneficial in situations that warrant it.

I think that negates the argument @SpudGunner makes, that he would have been better served by the More 2. It’s always best to have MORE (punny, eh?)

The screenshot of your Environmental Configuration shows how much Genie 2 manages all the components of your environment for various listening conditions.
Kind of a tailor made solution, based on your answers to the Pesonalization questions, and your RX.

@Volusiano said:

The other difference I notice between my P1 and the Speech in Noise program is the gain curves in the Fine Tuning section. I think Genie 2 boosts up my gain curve a little bit more for Speech in Noise.

Is it across the board, or limited to mid range frequencies?

In reply to @flashb1024

@flashb1024 said:

Is it across the board, or limited to mid range frequencies?

It seems to slightly raise the gain curve a little bit on the low end up to around 2 KHz in my case.

Below are the 2 screenshots for my P1 default and P2 Speech in Noise gain curves. It may be hard to tell because the increase is only slight. But if you’re in Genie 2 and you toggle between P1 and P2 with the same gain curve in place, then you do see the movement for the increase.

image
image

Edited

In reply to @flashb1024

@Volusiano said:

Difficult default to 6 dB in the default VAC+ P1 program, but if I add a Speech in Noise program, then Genie 2 sets the NNS Difficult default to 8 dB for Speech in Noise.

@flashb1024 said:

So, NNS is adjusted per listening conditions, and the extra reserve of the More 1 is beneficial in situations that warrant it.

I think that negates the argument @SpudGunner makes, that he would have been better served by the More 2. It’s always best to have MORE (punny, eh?)

Not really, I think @SpudGunner still makes a valid point because most people probably just stick with the default P1 program most of the times. That’s the whole premise of the Oticon OPN/S/More, that P1 can serve you just fine for the most part. Originally Oticon didn’t have Speech in Noise or Music or any of the other built-in programs. It was only through a perception by the public that the lack of other built-in programs makes the Oticon a less desirable brand did they start adding more programs almost a year after their introduction of the OPN.

Besides, I think Jim never said he would have been better served by the More 2. Anybody would be better served by the More 1 if money is no object. But in terms of the NNS, if I remember correctly, he said that he felt that the default max NNS seems adequate for him and maybe if he had the hindsight knowledge, he might have veered toward the More 2 for money saving.

But then when I mentioned the Spatial Sound differences, he got interested in what a spatial estimator is, and once he understood it better, he said that Spatial Sound LX would be an important feature to him to have 4 instead of 2 estimators. Jim, feel free to correct me if I remember anything wrong.

Personally, I think that the Speech in Noise program in the More should have had the 10 as the default instead of the 8 dB. If you have to remember to switch from the default P1 to the Speech in Noise program, that’s a clear intention through manual intervention you want the most NNS you can possibly get for your tier.

@Volusiano said:

Anybody would be better served by the More 1 if money is no object. But in terms of the NNS, if I remember correctly, he said that he felt that the default max NNS seems adequate for him and maybe if he had the hindsight knowledge, he might have veered toward the More 2 for money saving.

But then when I mentioned the Spatial Sound differences, he got interested in what a spatial estimator is, and once he understood it better, he said that Spatial Sound LX would be an important feature to him to have 4 instead of 2 estimators. Jim, feel free to correct me if I remember anything wrong.

No, @Volusiano, you get full marks for listening. In fact, those four spatial estimators that only More1s have are right up there with Clear Dynamics, in terms of making More1s worth the premium. The more I consider what they allow me to do, the closer they come to actually superseding Clear Dynamics’ value to me.

With a wedding with noisy situations, small gatherings, and more, I am very happy with improved speech recognition with the maxed out neural noise suppression, default DB levels in speech in noise, transient noise management, and slightly improved gains with enough reserve (5-7.5 DB left over).

I will continue to test and tweak the above settings such as Aware, etc as per the above images, once fitted later this week.

One thing I noticed is that my own voice sounds louder but to others, it sounds softer so I have to speak up louder than normal for them to hear me. I am generally known to be soft-spoken

@gkumar said:

One thing I noticed is that my own voice sounds louder but to others, it sounds softer so I have to speak up louder than normal for them to hear me. I am generally known to be soft-spoken

Is that through phone streaming? Or are you talking about you hearing yourself through the More louder, so it makes you speaks softer than normal so your own voice wouldn’t sound too loud on the More, prompting others to ask you to speak up louder?

If only through phone streaming, then can’t you just put your iPhone closer to your mouth? Or if using a ConnectClip, make sure its position is right under your chin via wearing the provided lanyard or clip it to your front pocket.

In reply to @Volusiano

Just FWIW, I have a client fitted with More1, who was initially experiencing some resolution issues. After a bit if tweakery and to make sure we were on target, I pulled that dynamic ‘difficulty’ slider all the way to the left.

In the office she thought they were ok, however in the real world there’s a complaint of harshness to a level of near discomfort. Now, I know all the Oticon marketing spiel says loads about how their is clever enough to raise the to +9dB, but from the outside it looks like everything is made louder (overprescribed) and if the AI is good enough to recognise noise sources, it might be able to strip these out.

The risk is, if you go into a coffeehouse environment, ‘all’ the speech is retained in a hugely complex sound field, while the occasional bit of steam frothing is cut out. I’m guessing you’d potentially leave with a bit of a headache; mitigated with the treble-tap on the iPhone home button.

She’s coming back in for a tweak down this week. I wasn’t expecting the More 1 to be ‘More Louder’ on the same gain settings. I ‘was’ expecting it to deal with the sound field better, in order that it picked out the speech (per the SNR claims) within the client’s preferred listening range.

That’s very true that if you’re in very complex environment where there are a ton of people around such that almost everything is speech, overwhelming anything else that may be noise, then it’s not good either because very little would be suppressed, and if the Sound Enhancer is set to Detail, that will raise the speech volume to even louder.

I wonder if in that situation, the user would turn on the MoreSound Booster in the ON app to narrow the directional beam forming to just the front only would be more helpful than using either the Speech in Noise or the default program P1?

I also wonder if just turning down the overall volume compensate for the increase in gain in all the voices helps, at least to reduce the discomfort of the very loud volume of all the cumulative voices.

Another thing is that if Oticon has trained 12 million sound scenes in the , I wonder how many of these sound scenes are of the type where you’re in a crowded bar or a coffee house, or the hallway of a theater during a break where everyone crowds out to the hallway and talk. And if they have included these scenarios in their 12 million sound scene, what the DNN outcome is like for them.

It’s one thing to be in a crowded restaurant where at least you’re still spaced out enough by table groupings to try manage separate group conversations. It’s another thing to be standing shoulder to shoulder with a sea of people in a tight space like a theater hallway during a break, where everyone is talking at the same time and the ceiling is extremely high and the echoes permeate and made worse by the high ceiling. Heck, I don’t even know how normal hearing manage those scenarios, but apparently they do OK because they’re all talking and listening and the conversations do carry on.

In iOS, you treble tap the home button and it takes you directly to the mic levels, programs and Live Streamer/Remote mic function without needing to find the App.

It’s like a ‘fast basic remote control’ screen.

Not seen an android equivalent.

Sorry, @Um_bongo - I was confusing treble with the treble in “treble clef” or the knobs on a guitar amp. When I Quacked it, “triple” kept coming up, and the lights went on.

[The language has gone to seed in the colonies!]

In reply to @gkumar

@gkumar said:

No this is in real life.

There’s a new feature in the Genie 2 2021.2.1 version called Fitting Assistance under Fitting → More Tools. Below is a screenshot which show various scenarios for your own voice and suggestions on how/what to adjust to help improve it.

It has other situations, too, beside your own voice with help to make adjustments. Also what’s nice is that you just click on either Right, Both, or Left to apply the recommended adjustment on the fly (if it’s not spelled out, click on the little “i” for Info to display it. There’s also an undo button. Sometimes they also have Additional Solutions as well.

image

We had a lot of discussion in another thread that was recently closed about NNS (and other stuff around the and also the differences between the More 1, 2, and 3 tiers) that really belongs more in this thread, and can be helpful to read if one is interested. So I just want to provide a link of that thread here in case anyone wants to read up on it. The relevant discussion on NNS started from thread # 346 (or 347?) with a post from @JeremyDC onward all the way to near the end of the thread before it got closed.

Going from Oticon OPNS1 to More1 aids
I stated in my op on the fluttering issue, that Oticon does have the ability to roll the firmware. They wanted me to return my ha’s to accomplish just that. My VA Chief of audiology offered me the option to be seen by a Community Provider who will have the same version of Genie, and hopefully resolve my programming issues. They certainly were not very happy that I downloaded the software. I doubt the availablity is a sign of generousity as much as a convenience for the dealer network. I’m…

I tried to have a tread that gave my experience with my aids, and as much my fault as anything the treat was basically taken away from me and taken over by everything but what I had in mind so I asked that it be closed. And I don’t regret doing so.

Oh, I’m sure you had your reasons, Chuck - we’re only saying that it’s too bad, that’s all.

Related discussions