Today, the help-in-noise features of hearing aids, they are so powerful that people are a little careful about actually going all in with these features, because they are so powerful. But by using ACT, we can identify the people who will really benefit from this. – Søren Laugesen, Ph.D., Research Manager, Interacoustics Research Unit
I think the key parameter in measuring speech in noise assistance is the SNR (signal to noise ratio) of the speech. If one doesn’t care at all about hearing any other sound except speech, then an excessive SNR might actually be preferrable to those people because the high SNR basically would suppress other sounds to almost nothing and all you can hear is speech. This is assuming that the method of achieving an excessively high SNR doesn’t cause a distortion in the speech signal.
For Oticon with their open paradigm, having a higher SNR to suppress the noise too aggressively is not consistent with their open paradigm. If you force Genie 2 to give you a very high Neural Noise Suppression by giving it a very high ACT number, or by forcing the DNN to give you artificially high SNR for speech via manual manipulation of the DNN parameters, you’ll end up with an unbalanced sound scene that would favor speech over any other sounds. If that’s what you prefer, then it’s fine, but you’ll lose out on the benefit of being able to hear other sounds well or clearly. That was the whole intent of the Oticon open paradigm.
The part I’m not clear about is when there’s no speech going on at all, whether forcing the Neural Noise Suppression to be too high to begin with will cause the sounds in the sound scene to be overly subdued unnecessarily or not. If yes, then this would be another adverse effect you don’t want to have.
Before, the traditional front beam forming approach was kind of the main way HA mfgs knew how to help boost the SNR on the speech. It’s like putting the blinders on a horse. But it’s better than nothing, at the expense of not hearing other sounds. But if you can find a different way where you can understand speech, but also be able to hear other sounds as well, wouldn’t you rather have it that way?
So I think as long as the approach to boost the SNR on speech doesn’t cause distortion on the speech, there’s really no adverse effect for doing it per se, as long as all you care about is to hear and understand speech, and you don’t want to hear anything else but speech. It becomes a matter of preference. But if you want to hear other surrounding sounds as well, as long as you can hear and understand speech clearly, then the adverse effect is that you lose out the ability to hear more sounds that you want to hear, just because you over tune the hearing aids to let you hear just speech only. After all, there are more sounds to life than just speech.
I appreciate your response, yet I find that I have trouble understanding it.
I’m caught…I was old when I got my first hearing aid 20+ years ago. And I didn’t replace that Widex with real hearing aids until I qualified for compensation due to loud noise at work. That was about 12 years ago…I had my first Phonaks for 7? years; my second Phonaks for 2 and now almost 2 years with the latest ones.
I’m having trouble hearing, and I’m trying to learn what to do to compensate.
I think the hypothetical topic of this thread only applies to people who are not still struggling with speech in noise anymore because the existing SNR for speech serves them adequately already. For folks who are still struggling with speech in noise, obviously the current SNR on speech is probably still not good enough for them, so to them, too much SNR is not an issue they have to worry about yet.
Your right.
Yet as my dispensing audi proved the quality of the hearing test is paramount. AND I can’t understand why speech in noise tests aren’t a priority in hearing aid testing.
Traditional hearing tests are the polar opposite of speech in noise testing. They put you in a “sound proof” booth…Force the sound directly into your ear via headset and pretend that they have done a comprehensive test.
The audiogram of most of the people on this forum would be far worse if it was done with a bit of background noise.
.02
Dan
You’ve just articulated my DREAM pair of aids. I’ve been pining away for that kind of focus for decades. And even tho the Phonak Lumity Life aids help with speech comprehension 85% of the time, it’s that last 15% when I’m in LOUD places and straining to get what’s being said even if I’m in a dedicated “Speech in Loud Noise” program.
It seems it’s rocket science to separate out the human speech frequencies from ambient noise. Maybe there are aids out now that use AI to help users selectively drill down to HUMAN speech, but that’s not what my Lumity Life aids do. When I put them in Speech in Loud Noise, ALL sound drops like a stone, volume down like 50-60%. So then comes the balancing act. I have to start cranking up the volume on my aids till the person facing me is at a volume where I may get 70% or so of what they’re saying, while keeping the ambient clatter down.
That’s as good as it gets for me. Even with the handheld/tabletop Roger mic, it picks up SO much ambient noise that I rarely use it in loud places. It is also incapable of drilling down to just HUMAN speech frequencies. No A.I. in those devices either, I guess. As you say, the beam-forming “blinders” on my aids are about all I can hope for.
LOL, that’s the truth! Boggles my mind how we go through all the rigamarole of tone tests and word recognition in a soundproof booth, get the aids back pre-programmed based on the results of those narrow-minded tests, and then find that LIFE - outside the booth! - is far more complicated. No audis even test us once again with the aids IN and tones or words flying. Kinda funny, that?