Oticon Intent review at 3 weeks

Thank you @eliotb for you detailed Intent 1 review. I assume that it’s the Intent 1 and not the 2, 3 or 4, right?

Do you know what your MoreSound Intelligence Neural Noise Suppression max settings are for Easy and Difficult Environment? I just wonder if it’s set to the max possible value of 12 dB for Difficult or not?

It’s an Intent 1 … and where would I find those settings? Do I need to ask my audiologist what’s in her software, or can I find it buried within the app?

The Xceed uses the OpenSound Navigator, which is the noise reduction technology for the OPN and the OPN S. It does noise reduction differently than the More/Real/Intent. The OpenSound Navigator cleans up the speech diffused with noise by creating a noise model of sounds from the sides and rear and subtract this noise model from the speech in front that is diffused with the noise (that is represented in the noise model). This helps give a clearer speech that is minimally diffused with noise.

The More/Real/Intent uses the AI DNN (deep neural network) to break down sound components from the sound scene and recreates it. And in the process of recreating it, it can rebalance the sound components to give speech more volume and non-speech sounds less volume as a way of achieving noise reduction and still be consistent with the Oticon open paradigm.

Initial reviews on the Intent seems to be positive on speech in noise compared to the Real. I think this is 2-fold: 1) Oticon is able to eke out an additional 2 dB of SNR for speech, which hopefully will make speech stand out even more, and 2) the DNN 2.0 is trained with more diverse sound samples, which hopefully improves the accuracy of the sounds characteristics (sounds seem more real).

Having explained the 2 approaches (of the OpenSound Navigator or OSN in the Xceed and OPN/S vs More/Real/Intent), it’s more obvious that the OSN “cleans up” the speech from diffused noise, while the DNN rebuilds the speech components so that the emulated speech is “originally” clean already.

Now back to your question regarding max NR, whether it’s always good to set it to the maximum available value or not. After all, you pay a premium for the top tier aid, why not take advantage of the best NR you can get? Evidently if it’s always good, then Oticon would not have given you a choice of setting what the max NR is, but instead would just always apply the max regardless. But obviously there is a trade-off in scrubbing the diffused noise from the speech, because if you scrub too much of the diffused noise, you’ll scrub into the speech components that are masked by this diffused noise as well. Kind of like scrubbing bugs off a car finish. You get too abrasive, you might get through the clear coat and maybe even down to the paint.

So I’m guessing this is the effect that you see with the Xceed. If you have it set at -9 dB NR max, it may scrub too much diffused noise off that you lose some integrity of the speech component that gets scrubbed along with the noise. If you set it at -7 dB NR max, you don’t scrub off the noise as much, but at least the integrity of the speech might not be compromised as much. There is a trade-off here depending on your tolerance of the noise level, to be balanced with preserving the speech integrity. And that trade-off is left for your HCP and you to decide.

It’s also worth clarifying that the max NR setting Oticon gives us is not the “always on” flat out NR that is applied for any situation. Oticon has a way to determine the complexity of the environment in real time, and applies the NR only proportionally to the actual level of complexity at the moment. For example, if you set your max NR to be -5 dB and you have 6 scenarios of increasing complexity, S1 being the simplest to S6 being the most complex. Then if -5 dB max NR is what you set, then S1 is given 0 dB, S2 given -1 dB, S3 given -2 dB, so on until S6 is given -5 dB. But if you had set your max NR to -9 dB, then S1 would be given -1.5 dB, S2 given -3 dB, S3 give -4.5 dB, S4 given -6 dB, S5 given -7.5 dB, and S6 given -9 dB.

Of course this is just an example based on my personal guess, and perhaps it’s not that linear like I presume it to be. But at least it gives an idea of scale, to point out that Oticon applies only the appropriate NR to the appropriate complexity level, and Oticon gives you a crude option to put a ceiling on what the max amount of NR you want. This goes back to how well you can tolerate the noise. If you can tolerate it better, you set a lower max NR limit so Oticon won’t scratch too deep as to possibly mar the speech surface (so to speak). But if you cannot tolerate the noise well, then you have choose to have Oticon do deeper scrubbing of the noise, but at the expense of marring the speech surface too much and lose some integrity of the speech sound.

Hello,

I have been playing very often with difficult situations between max noise reduction and normal noise reduction.
Background stories
1)Last December, I sent my MORE 1 hearing aids to OTICON for the mic replacement. And then I had used Oticon REAL 1 for over a month.
But, later I found that Oticon INTENT is coming…
So I quickly contacted to my hearing center and he accepted the return. And then when INTENT is available, he will send me w/ INTENT. In the meantime, I used OTICON MORE which was fixed.
2)However, my cutty dog chewed my hearing aids, OTICON MORE 1. When I send the broken hearing aids to Oticon, I was told that my MORE 1 was officially dead and no way to fix.
3)Thankfully, my friend in the States could send me his OTICON REAL1 because he just got OTICON INTENT 1. Now I am using my friend’s OTICON MORE 1 until getting OTICON INTENT.

For Oticon MORE1, when I maximized noise reduction, I could feel the comfortable sound but I misheard a few alphabets, especially the person speaking very softly such as 1955/1965 or Fisher/fiitcher. There is no problem understanding the sentence because we understand the meaning through intonation/context. That’s why I couldn’t maximize the noise reduction for difficult situations. I kept using w/ 8dB (medium reduction which Genie recommended)

However, For REAL1, there is a way I can control it in the app by increasing the dB (EQ) for mid-range which is mainly focusing on speaking. I tested and I found the best combination.
"Maximize 10 dB for Difficult situations -noise reduction" + Only When I work, I increase the mid-range by 3 dB/High-range by 1dB and decrease the low range by 2 dB.

Right now I am quite happy with REAL 1 setting and just satisfied with the setting. I feel like it’s more likely self-satisfaction that I can utilize full functions without any sacrifice.

As long as you can control the EQ on the app, you can add a little bit of clarity to maximize the noise reduction.

Thanks for this. As someone who has had the Real 1s only since the summer, I too wonder whether this might be one of those exceptional cases where it’s worth upgrading now and not waiting to skip a generation or two.

In his “review” of the Intent, Cliff Olson stated that the improvements are so significant that the “skip a generation or two” upgrading rule might not apply.

Furthermore, because I “only” have moderate hearing loss, understanding speech in noise is my primary reason for having hearing aids.

It seems worth it to pay $125 for a trial.

Thanks for noting domes/molds might affect how a hearing aid is perceived. The envelope I have describes them as “Bass Domes 6mm Double Vent.”

And greetings to you, Penzance99. Vienna is one of my favorite cities in the world from all my travels. I spent substantial time there courtesy of IAEA work.

Greetings from Vienna to the US - I also initially had double vented bass domes, but they were too occlusive for me. I then discovered the open bass domes, which work much better for me. But I guess the price which I have to pay for the open bass domes is that speech in noise performance is not always ideal, presumably because the noise reduction cannot work fully. So I am just wondering whether I can expect a huge improvement for speech in noise with the Intents as compared to the Reals under these circumstances?

There’s only one way to find out, i.e. test drive Intent with some of your existing domes on the end. I ended up with these years ago because the 8mm ones (perhaps closed, I don’t recall) made my ears feel “stuffy,” if that’s a proper descriptor. These 6mm ones seem to work for me, so I’d suggest testing Intent with both types if you can and see what happens. Best of luck. If it works, I’d suggest a celebratory shokolade mit schlagobers.

@Volusiano said:

The More/Real/Intent uses the AI DNN (deep neural network) to break down sound components from the sound scene and recreates it. And in the process of recreating it, it can rebalance the sound components to give speech more volume and non-speech sounds less volume as a way of achieving noise reduction and still be consistent with the Oticon open paradigm.

That’s a very good and concise explanation.Thank you!

Happily, speech in noise is NOT one of my issues, so the Intents don’t seem appropriate for me. I would have thought that breaking down sound components and rebuilding them would result in a more artificial sound, but no one has mentioned this. It’s a very interesting and new (?) approach to processing sound.

Can I assume that when the “speech in noise” feature is not engaged, the aids work like other aids?

I don’t see how it has anything to do with “deep neural networks”, but hey, advertisers gotta advertise and “neural networks” sound contemporary and science-y.

@1Bluejay said:

It seems all my life I’ve just wanted to boost HUMAN speech and be able to hear conversation out of the babble of noise (ambient as well as crowd-related)

It’s an incredibly difficult problem for hearing aid manufacturers to solve.

At any rate, I’m on day two of my trial with the Oticon Intent 2 with power domes, and I’m fairly blown away by how good speech-in-noise is. I had the windows down in the car while driving my wife to work, and I could hear her easily, comfortably, and with a stunning level of clarity. Almost as if the wind noise didn’t exist. I didn’t have to lean in. I didn’t have to focus at all on trying to understand her. I just heard her words.

My previous hearing aids were Starkey Halo 2 i2400 from 2016 (8 years ago) with 7mm open domes. I probably never should have worn open domes in the first place with my moderate-to-moderately severe hearing loss, so maybe that had something to do with it. A change in audiologist put me in proper domes. My current audiologist initially had me in double-vented domes, but my ears can’t fit 8mm and 6mm are too small, so we went with power domes. I don’t actually like power domes, but that’s something I’ll take up with her in a few weeks.

Tonight I went to Chick-fil-A to pick up a mobile order. It was super busy as usual, and yet when the employee called out my name to tell me my order was ready, it was like he was speaking directly to me. I really couldn’t believe it. It was like he was shining a spotlight of sound directly into my ears, while the voices around me receded into the background. Maybe not quite that dramatic, but I was just happy how effortlessly I could hear him. I’ve had to rely on my wife to interpret interactions with cashiers and store employees for a while now. Tonight, with my Oticon Intent 2, I didn’t have to.

My audiologist also used real-ear measurements to program my hearing aids (even called it “best practices” like Dr. Cliff on YouTube). No previous audiologist in my 20+ years of wearing hearing aids had performed real-ear measurements on me. That also might have something to do with my impressions here.

@jeffrey said:

Can I assume that when the “speech in noise” feature is not engaged, the aids work like other aids?

The Oticon aids, starting from the OPN onward (to OPN S, More, Real, and now Intent) always dynamically (constantly) scans and analyzes the sound scene 500 times per second to correctly map the different sound sources. Then based on this scanning and mapping, it calculates the optimal signal to noise ratio (SNR) of the sound sources, and the optimal noise levels (in 24 independent channels) to determine the amount of noise suppression it needs to apply. Generally, the higher the noise level, and the lower the SNR, the most noise suppression is given to a sound source (as seen in the screenshot below).

So the Oticon aids are DIFFERENT than other brands’ aids that usually determine how much noise there is in the environment through some sort of “Auto Sense” mechanism, then “engage” in noise suppression by switching automatically from the default normal program to a different program specifically built for speech in noise. And usually, noise suppression is by using directional beam forming for the most part.

Instead, the Oticon aids constantly does the analysis of the sound scene 500 times a second and applies the right amount of noise suppression help to speech instantly. However, the applied amount of help, once set, does not change unless persistent changes to the environment last for at least 2 seconds, to provide for some stability. And the noise suppression is applied using not just regular beam forming (but the MVDR kind instead), but also cleaning up the diffused noise that is still stuck with the speech with the OPN & OPN S, or with the DNN in the More/Real/Intent to boost the speech volume and lower the other sounds’ volumes.

So with Oticon aids, unlike what you think, there is no “engaging” the speech in noise feature by switching back and forth between the normal program and the speech-in-noise program as done by some kind of Auto Sense feature. The Oticon aids in the default program already ALWAYS have the appropriate level of noise reduction help determined and assigned, and will constantly adjust every 2 seconds if changes in environment is persistent for more than 2 seconds. There is actually no need to switch to a built-in speech in noise program in order to engage the speech in noise “feature”.

In the beginning, Oticon only released a single default Program 1 and didn’t offer any other built-in programs (like Speech in Noise or Music or Comfort, etc). The idea was that the default program already can do noise reduction in itself, so why need a speech in noise program? But the public didn’t understand and perceived the new Oticon aid at the time (the OPN) to be lacking compared to other brands. So about a year later, Oticon caved in and released built-in programs to match with other brands’ availability. But this is only a marketing things. In reality, the default program, if set with the same parameter values as the built-in speech in noise program, can handle anywhere from simple to complex environments just like the speech in noise program, because the noise reduction help is already assigned at the appropriate level automatically all the times, in the single default program.

@jeffrey said:

I don’t see how it has anything to do with “deep neural networks”, but hey, advertisers gotta advertise and “neural networks” sound contemporary and science-y.

Non-technical folks may not understand how DNN works, but it doesn’t mean that Oticon is making it up just as a marketing ploy to deceive people into buying their products. That would be advertisement fraud in a major scale. For technical folks, the use of DNN is very real and it’s happening more and more in many applications in life.

The use of DNN allows things that deem impossible to do before to become a possibility now. Specifically for Oticon, they first started with the OPN in 2016 and wanted to break the mold of frontal beam forming by introducing the open paradigm. This philosophy is kinda like “you can have the cake and eat it, too”. Now you can hear most of the sounds around you, yet still understand speech without having to put on the frontal beam forming blinders on to just hear speech in front but nothing else. They came up with a way to clean up the noise diffused in the speech by creating a noise model (of sounds in the back and the sides), then subtract it from the speech. It’s similar to the concept used in noise cancelling headphones. It’s easy to do in headphones because the noise in the environment and the streaming content are 2 separate sources. It’s harder to do with hearing aids because they come through from the same mics.

Anyway, this allows Oticon to make speech (no longer diffused with noise) clearer to hear amongst other surrounding sounds, without having to block out surrounding sounds, which is how they enable their open paradigm to work.

The DNN in the More through the Intent now does it in a different and more clever way. There is no more need to create noise models to subtract from speech anymore, because now sound components are processed and separated and recreated with their own adjustable volume levels to create a new balance sound scenes based on the parameters specified by the users.

This is not the place to explain how DNN works at a more detailed level. There are YouTube videos that can explain it very well, although it still requires a certain level of technical proficiency to understand it. But so many advances in our society have been possible and have enriched our lives because of math and science, and I can tell you that the concept of building a DNN network is heavily relying on the math and science principles that we’ve learned so far. So it’s not just marketing gobbledygook aimed to defraud consumers.

@icanseeyourhalo said:

It was like he was shining a spotlight of sound directly into my ears, while the voices around me receded into the background.

Well, you pretty much described exactly what I’ve been pining away for these many decades. THAT kind of focus. Exactly on what the priority is: someone talking to you, not the ambient sounds.

I find that in noisy places, my Phonak aids tend to take the volume of EVERYTHING down - too much. That’s with Noise Management set to its least aggressive setting. If I put the aids into Speech in Loud Noise, the effect is (comically?) even MORE - like on steroids. So I typically put the aids into that program and immediately jack the volume up two steps. Maybe it’s a mental thing, but I find if I then increase the volume, I can somehow pick out what’s being said but that doesn’t mean it happens EASILY.

So that’s what is prompting me to look into the Oticon Intent. I guess my other critical necessities would be:

• Battery life between charges (vs 17 hrs/day with Phonak rechargeable)
• Ease of streaming to phone, TV, laptop
• Use of external accessory mic so I can stream with TVs on the road - places where I may not be able to plug a TV Connector into both TV and the power socket cuz they are located too far apart or it’s too tight a squeeze to get at.
• Other useful programs, two being: Music (no messing with dynamic range) and a program to let me hear on the phone stereophonically like I do with Acoustic Phone on my Phonaks on devices that my aids are not paired + connected to.
• Overall natural sound quality is up there, but not my #1 criteria.

I feel like a traitor to myself cuz what I really want is a BATTERY operated aid. I’m so done with the rechargeable cludge, but I highly doubt Oticon will come out with something like that. In fact, it seems that ALL the new aids coming out are rechargeable nowdays - no choice in that anymore. If Oticon delivered on the speech in LOUD noise promise, that may be the deciding factor.