Skip to content
HearingTracker
Log in

Cannot understand voices in supermarket

This topic is closed.

29 replies2.4k viewsLast reply

Jump
Link to this topic
@SpudGunner said:

What would make Phonak better than, say, Oticon or Widex, in this regard?

Phonak can move sounds to lower frequencies than other brands. At least this is my understanding.
Pretty amazing technology, whatever brand is used.

Yes, it is. Very effective, from what my experience is. I’m not sure if the lower floor for transposition would be useful for me.

Last tweaking session, my audi moved my transposition band up from a lower setting @Volusiano had recommended. I haven’t noticed anything different, to be honest.

It’s interesting, though, and highly prone to hyping (byall makers), IMO.

@SpudGunner said:

I’m not sure if the lower floor for transposition would be useful for me.

Your high frequency hearing loss not that bad at this time. If you can actually hear the upper frequencies you don’t have to use technology. But there might be a time it really helps you. An example is using a less powerful hearing aid or receiver while using frequency lowering technology, typically ski slope hearing losses.

The OP may benefit from frequency lowering technology, need to see their .

The frequency compression from SoundRecover 2 definitely has the advantage of having the lowest destination band compared to the frequency transposition/composition of Speech Rescue. For some folks where their ski slope happens much earlier on in the frequency range, SoundRecover 2 can be more helpful than Speech Rescue for sure.

For folks where needing the lowest destination band is not really an issue, Speech Rescue has its advantages. It can keep the high frequency amplification as is, so the is added ON TOP of it, so you get to hear both the original sound amplified as is, and the lowered sound to help give it a more audible boost. On the other hand, once SoundRecover2 compresses the high frequency sound, you lose the original sound for good because it gets squished together for good.

The other advantage of frequency transposition vs frequency compression is that the source frequency band get transposed to a lowered band, but don’t get squished into a smaller range like with frequency compression. In theory this may seem better, but in practice, once the user adapts and gets used to how the new lowered sounds sound like, then what was not natural eventually becomes natural anyway.

The practice of frequency composition also allows a much larger source band be cut up into 3 equal length bands to be superimposed on top of each other in the destination band. This slicing and dicing and stacking avoids further “squishing” of the frequency compression. But again, this may help things sound more natural in the beginning, but users of SoundRecover2 can adapt the compression to sound natural to them eventually anyway.

I don’t think Oticon stuck to frequency transposition and composition because the compression got patented in the beginning. I think they believe they have a better approach on paper (and probably eventually also proven in practice), so they stuck with it. I think they can easily move the destination band lower to make it competitive with SoundRecover 2 if that’s what they want, they don’t need to resort to compression just to lower their destination band. After all, it’s just digital signal processing and manipulation to them. But there’s probably a price to pay to lower the destination band to too low and it’d probably start interfere with other more important audible cues, so they probably prefer to stick to their lowest band as is. Still, I wish they just make even lower destination bands available and let the user choose for themselves.

Edited

In reply to @SpudGunner

@SpudGunner said:

Yes, it is. Very effective, from what my experience is. I’m not sure if the lower floor for transposition would be useful for me.

Last tweaking session, my audi moved my transposition band up from a lower setting @Volusiano had recommended. I haven’t noticed anything different, to be honest.

It’s interesting, though, and highly prone to hyping (byall makers), IMO.

From looking at your , Jim, you can probably set your destination band to as high as 3 KHz on the left and 4 KHz on the right, and still have quite audible lowered sound.

Yeah, as long as it is in your audible range, then selection of the destination band should not make much different in how it sounds, because “s” and “sh” will still sound like “s” and “sh” whether they are a little lower or higher frequency. It’s not so much about how/if they sound different, it’s about whether the sound is audible to you or not.

If you look at the screenshot below, the left chart (red) has been set to the lowest destination band of 2.4 (the red column), and as you can see, that is where I have the most amplification and where it should be the most audible to me.

On the right chart, it’s been set to the highest destination band of 5.5 (the blue column), and as you can see, that is where I have the worst hearing loss and need the most amplification on, yet the gain curve there looks the weakest in that blue column. So the lowered sound in that blue destination band would not be very audible to me. Although I can increase the Strength setting on the right side to offset the lowered sound and make it more audible, it’d still be better if I had chosen one of the lower destination bands toward the left side in the first place because the lowered sounds would be more audible there.

So if I had chosen 2.4, or 2.7, or 3.0, or 3.2, or even 3.5, like shown in the second screenshot below, it’d still be adequately audible to me through the amplification in that range, although the audibility is probably weaker in the 3.5 configuration as compared to the 2.4 configuration.

So it’s a game of picking the configuration where you like the lowered sound the most. If you think the sound at 3.5 is the best, then pick 3.5 as long as it’s still adequately audible. But if they all sound the same between 2.4 through 3.5, and 2.4 is the most audible to you, then you’d probably want to go with 2.4.

If you have your audi do an in-situ audiometry in Genie 2 for you then re-prescribe based on your in-situ result instead of the audiogram result, Genie 2 will give you an option to re-prescribe the Speech Rescue configuration for you as well, and it’ll give you its recommended configuration that it thinks is best suited for you.

image

image

@Volusiano: IIRC, my audiologist moved me up from 2.4 to 2.7 - he said something about 2.4 being pretty far from my prescribed level. I’m getting to trust him a fair bit, and - seeing as my technical understanding is so far below yours - I just took his word for it and didn’t ask questions.

I haven’t had much time to contemplate the settings changes this week. All I know is that I can hear pretty well, all things considered.

[I think that I’m hearing more “sh” coming out of “s” when I’m streaming, but not with live voices. (Some would say"full of ‘sh’", perhaps? I dunno.)]

Yeah, moving from 2.4 to 2.7 is really not much at all. If you look at your below between 1.5 KHz to 3 KHz, the slope of the loss is fairly flat in that area, so the audibility should be almost the same and really there’s no concern regarding audibility with this small change.

image

Phonak’s frequency compression is also adaptive–it lowers high frequency phonemes and then releases the compression on low frequency phonemes to maintain other acoustic cues.

Audibility of the /s/ can be confirmed with so you don’t need to guess at what the appropriate settings are. This has been true for my entire career. It is very common for results on verification to be very different than results ‘suggested’ by the software in either direction (e.g. what the software thinks is audible is not, or what the software thinks is inaudible is).

Related discussions