In looking for advice on molds and venting, I came across this curious but rather long YouTube video where a provider attempts to instruct other providers on the A, B, C’s of fitting and how to deal with typical special fitting problems. His YouTube video certainly doesn’t have many subscribers, Likes, or followers. But it’s interesting in the way he attempts to explain all from one complex diagram, e.g., how to deal with soft-spoken people, tinniness, feedback, etc. Interesting that he draws a line at 1.5 KHz and says he pays most attention to what’s happening with lower frequency sounds and the higher frequency sounds for the most part “take care of themselves” through gross adjustment to gain in the fit curve to deal with problems related to lower frequency sounds.
I’m an ignorant newbie who should be doing other things. But I thought some users might find some helpful ideas to discuss with their provider if they see themselves in any fitting issues the guy discusses and experienced users or providers might have fun critiquing what the guy got right or wrong.
If you watch it on YouTube, you can “Open Transcript” from the three dot menu on the right just below the video window and using Control +F search within the transcript for key words like “soft” - which I did to find his advice for Noreen on dealing with soft-spoken people (occurs at about 13 min into video).
My TLDR version of this is that he’s talking about determing what power of receiver to use. His advice is similar from what I got from a Connexx 7 instruction video–Look at lower frequencies (under 1500) on the fitting curve to determine receiver power. Don’t fret about big losses in higher frequencies to determine frequency power. He also suggests cutoffs for open vs closed fittings, but I’m really beginning to think that this is more of a personal preference thing. If you hate the occlusion, go to a more . If you want more clarity or better noise reduction, consider a more closed fit. Heck, you (Jim) are in a custom mold and happy with it and guidelines would suggest open domes.
It is an interesting video on the subjective, lets say, seat of the pants, approach to fitting. He does make a lot of good points in my amateur opinion. I am not sure what ’s you have but out of curiosity I did a quick transfer of your to Connexx and selected a Rexton Emerald 80 8C HA (Kirkland Signature 8.0), and I would say you are a prefect fit to a S receiver with an open fitting. There should be no need for molds. I would pick a vented click sleeve. Here is what it looks like:
The fitting curve would look something like this for using vented sleeves.
@mdb and @Sierra - Great advice you both give. I’m with you MDB on molds vs. domes and venting. From my own experience with the variability of my ear canals with domes I had no control over the degree of venting and just as you say, the increased control over clarity and noise that comes with molds is great. Introducing any occlusion into a fitting when not absolutely necessary seems to be the great bogeyman of providers re patient satisfaction and avoiding returns…
Impressive, Sierra, that you input my into your fitting software. Before I got my ’s (Quattro’s) I realized the Low Power receivers would fit my loss but I specifically requested the Medium Power from my audi. At the time I just wanted a more powerful receiver in case my hearing loss worsened during the 5 to 6 years that I wanted to make my HA’s last because of the expense. But now that I (think that I) know more about how HA’s work, it’s my understanding that a more powerful receiver gives me more hearing “space” to compress sounds into - there can be a wider range of amplitude to put frequencies that I can’t hear without an aid into so that the differences between soft, medium, and loud sounds can be greater.
I don’t think the ReSound Smart Fit software shows graphically the effect of wearing double domes or a mold in the way the Harmony II software used by the YouTube’er or the Connex software does. That’s neat for clearly showing that control of venting mainly works at low frequencies and I wished Smart Fit could do that.
On the guy’s presentation, I guess I mainly found it interesting for his 1.5 KHz demarcation line, his comments on venting, and dealing with special problems and the one diagram that he tried to use to explain it all plus his implied “the high frequencies take care of themselves” when selecting the right power receiver for low frequency loss. As Sierra says, it was the guy’s “seat of the pants” approach to fitting all condensed into one diagram that gets one’s attention and gives one something concrete to look at and critique.
I think your issue would be in using a M receiver when it isn’t necessary is, as the guy in the video says, the inability to turn the gain down low enough in the 250 to 1K range. This is how it fits on the Rexton M receiver. That said, we have trouble hearing differences of 3 of less, so the effect may be negligible in practical terms.
I’m not sure I buy “the highs take care of themselves.” I think it’s more along the lines of, at least for losses like mine, their’s not much you can do about it with more gain, so don’t worry about it. If you want the information the highs provide, of some sort is an option. My “gut” with only my own experience to back it up is that higher powered receivers have more floor noise. I’m quite happy with an S receiver in my KS7s. A medium receiver in some Oticon Rias was awful for noise and a M in Signia NX was ok, but had detectable floor noise. A lot of this stuff is along the lines of whatever works. They’re only guidelines.
I think your issue would be in using a M receiver when it isn’t necessary is, as the guy in the video says, the inability to turn the gain down low enough in the 250 to 1K range.
ReSound receivers seem to behave differently. As one goes up through low, medium, high, and ultra power, there is no difference in minimum fitting range levels vs. frequency until one gets to ultra power. All the changes in receiver gain performance are in maximum possible amplification and, of course, there are differences in background noise and harmonic distortion, etc. I might have more to say later when I understand better the implications of the top part of the fitting range graph (20 range). Also for my fit as a first-time user, no gain was applied to 50, 65, or 80 dB input until I was above 1.5 KHz. If I change to Experienced (Non-Linear) gain is only applied across soft, medium, and loud input at 1.5 KHz. So for my hearing loss, no gain is prescribed/needed in low frequencies following the + proprietary ReSound (comfort) fit. Things would change drastically if I switched to DSL5-Adult, which goes in for a lot more amplification. So fitting algorithm to be used is very important in deciding on fitting range, too.
P.S. My ignorant understanding of the top part of a fitting range (around the 20 dB level) is that it’s more defined by processor noise level than anything else, e.g., if the processor noise level is 20 dB, you’ll never be able to hear (well) at 10 or 0 dB. The YouTube’er’s comment about inherent over-amplification at low frequencies the greater the receiver power, I did not understand in terms of the upper part of the range curve.
ReSound Low Power Receiver Range ReSound Medium Power Receiver Range
I have spent a fair bit of time comparing v5 and fitting formulas. While there are some fairly small difference in compression they prescribe, I think the main difference is the high frequencies. The DSL v5 does try to amplify them a lot more. I ran the simulation again with DSL v5 and you can see the higher gain it is applying at high frequencies. I am not sure of the implications of the indicated negative gain at 250 Hz. It would seem that the amplifier has no way of providing negative gain…
I don’t know what to make of the negative gain, although it might be possible with a closed fitting. I’m not saying this actually happens, but it seems feasible that with a closed fitting, a hearing aid could detect a 80dB sound at 250dB and send a 65dB sound to the receiver.
While there are some fairly small difference in compression they prescribe, I think the main difference is the high frequencies
We’re getting away a bit from critiquing the YouTuber’s comments on power and venting and related fitting problems but you’re right on DSL5-Adult vs. . I’ve examined them for my fit vs. the ReSound +, which ReSound says is more akin to the original NAL-NL(1). ReSound claims its users actually have a better speech-in-noise experience with Audiogram+ but I’ve briefly tried both NAL-NL2 and DSL5-Adult and think that they both have at least a slight edge over Audiogram+ in speech clarity. They both bring up soft input compared to Audiogram+. But DSL5 really cranks up the high frequencies and shows no mercy on loud input. The thing about NAL-NL2 is it backs off gain for loud input (80 ). Consequently, it has far more compression than either Audiogram+ or DSL5-Adult. DSL5 is way too loud for me. Running aerated faucet water is very annoying. With NAL-NL2, although soft sounds are a bit more aggressive than Audiogram+, some of the loud sounds are actually more restrained - so it might be a keeper. The price for “best” speech clarity, though, might be a less natural sound environment for average listening than a fitting algorithm that is flatter and more generously spreads relative sound volume out rather than compressing it all down to a more narrow fitting range.
Here’s a good article comparing NAL-NL2 with DSLv5
Thanks for the reference. I came across it, too, in preparing to try NAL-NL2 vs. DSL5-adult early this AM. I have seen other material that says that the exact way a OEM implements NAL-NL2 or DSL5-Adult can affect the fitting outcome for an individual (forget whether article you reference mentions that consideration). So apparently not all implementations of these generic protocols are created equal.
On how the fitting range is not the be-all, end-all for whether one will have a satisfactory fit for a hearing loss, the following article discusses some other considerations such as fitting algorithm to be used, dome and venting. Fig. 3c shows that a fitting range that was O.K. for a bass dome, single vent is not OK for an open vented dome, etc. Fig. 1 compares fits for OEM’s comfort fit, NAL-NL2, and DSL5-Adult. My own results for MP receivers for Audiogram+, NAL-NL2, and DSL5 are shown in the pictures below, as predicted by ReSound’s Smart Fit 1.4. And a last picture for DSL5 with a Low Power Receiver. The pictures shown that even a Medium Power receiver barely handles the high frequency amplification required for my loss according to DSL5 and a Low Power receiver tops out and the amplification is clipped at the highest frequencies. The inability of the Low Power receiver to handle loud input for the highest frequencies creates, in effect, an artificial unprescribed sound compression (see output table for Medium vs Low Power fits below).
Another useful thing that I learned along the way is that age, sex, vs. a monoaural fit are important to determine the prescribed gains for the various fitting algorithms. I hadn’t been putting age and sex into my fitting software before calculating prescribed fit but from now on I shall be sure to do so (a binaural fit, for example, reduces the prescribed fit for each ear about 3 dB as compared to a monaural fit, if I remember correctly - think it’s for NAL-NL2).
Update_Correction: The binaural fitting difference vs. monoaural out of the paper MDB cites in his post above is: “Bilateral fittings: DSL m[i/o] prescribed targets for speech are reduced by 3 dB across input levels for bilateral fittings re: unilateral fittings (Scollie et al., 2005). NAL-NL2 has a bilateral gain correction that increases with input level. The correction for binaural summation is 2 dB at low input levels and up to 6 dB at high input levels for symmetric losses, and less for asymmetric losses.” End_Correction
In the images below, the black dashed lines are the prescribed output gain for 50, 65, and 80 input, the red boxes are the actual output the software predicts the receiver will realize in the “average” ear. So on the right of the DSL5 Low Power Receiver (last image), one can see that this receiver is incapable, by a few dB, of producing the prescribed output at high frequencies and in consequence moderate (65 dB) and loud (80 dB) input at some high frequencies are squashed together and have ~the same volume to the ear. The solid red line at the top of each curve is the Maximum Power Output of the receiver.
Audiogram+(Medium Power Receiver) NAL-NL2(Medium Power Receiver)
DSL5-Adult (Medium Power Receiver ) DSL5-Adult(Low Power Receiver)
But DSL5 really cranks up the high frequencies and shows no mercy on loud input. The thing about NAL-NL2 is it backs off gain for loud input (80 dB).
Yes sir, it does, and also cranked up my . I lowered MPO and that helped a lot. I like DSL. Things are clear and it seems to be handling high Sound Recover (frequency compression) settings better that the default Phonak formula.
Yes, your comments about DSL5-Adult and MDB’s about motivated me to try something other than ReSound’s proprietary fit. Reading your older remarks about molds was part of my motivation to get them and go with ~1 mm vent via select-a-vent. I like the noise control and the sound clarity effect afforded by molds with narrow vent, don’t mind a slight occlusion. Your latest posts on your experiments seem to say that you lowered the by 9 , raised soft sounds by 2 dB, and (maybe, lowered loud sounds by 1 dB?). My ignorance shows in that I don’t understand if lowering MPO frequency by frequency automatically lowers all gain injections by a proportional amount based on a calculated predicted compression ratio for each frequency in your programming software? The other thing I was wondering, since you are using Sound Recover and lowering (compressing) high frequencies, was whether that makes the “high” frequencies for you actually much more tolerable than they would be for me as I am not using any frequency shifting. So maybe what’s quite OK for you shifted to a lower frequency makes me want to climb the walls when presented at a unmodified noticeably higher frequency? Perhaps also by frequency shifting, you have more amplification room to work with your “high” frequencies whereas by me keeping my high frequencies up in a region of narrowed remaining hearing, my high frequencies at soft, moderate, and loud are all getting more squished together in a narrower range of possible amplification?
I want to experience NAL-NL2 for a while before giving DSL5-Adult another shot with your recipe applied as best that I can. Then I’ll be better able to appreciate any differences between the two. Right now I hear the wife great but she’s not as eager as I am to rush out to Chili’s or similar for the noisy restaurant test!.
It would be interesting to see in target output diagrams such as I posted how your present modified DSL setup is different than it would be if you were using NAL-NL2. It almost seems like you’ve changed DSL5 to be more like NAL-NL2 in lowering permissible MPO and permissible loudness. Maybe your current soft sounds are louder than they would be by either canonical recipe for your hearing loss.
Maybe your current soft sounds are louder than they would be by either canonical recipe for your hearing loss.
Yes, well, sort of. The approach seems to be, in part, to greatly increase the highs. The mid to low frequencies were actually lower with DSL than either or . So I did raise soft speech (the 50db line) across all frequencies.
I didn’t bother adjusting the high frequencies but it doesn’t help me to have high frequencies blasting away because of my dead spots.
DSL has less compression, which always makes things sound clearer to me. Is that true for everyone? The APD compression ratios were in the 2.2 to 3.1 range, and DSL ended up with those in the 1.6 to 2.0 range. I’ve had hearing aids for 9 years now and I don’t normally have a problem with loud sounds but I remember liking the lower compression ratios even when I got the Resound Aleras in 2011.
So then, with the DSL fitting formula in place and soft sounds bumped up, and things sounding clear, I cranked up the Sound Recover 2 . I wanted to go as far as possible with that as long as it increases audibility. I started off starting that at 800 and 900 Hz and made one adjustment so now I’m at 900 and 1100. That’s pretty extreme and I’m watching out for/ S/ and /SH/ problems. My goal with frequency lowering is to get as much as possible off the dead spots in the 4k - 5k region, and increase audibility.
It is funny to me though, music sounds great. How is this possible?
Anyway, I like DSL and will probably stick with it but you can do the same thing with any fitting formula, do the basic first fit and then a couple of adjustments you know you need. I do think the the company fitting formulas are tweaked for comfort in order to reduce returns and NL2 and DSL are strictly for audibility, but you can adjust the company formulas to accomplish the same thing.
Video is pretty good. How quickly you can close up the vent will vary from person to person, and how quickly you can move into a more powerful receiver also seems to vary from brand to brand. Some manufacturer power receivers are just not as noisy as others.
DSL has less compression, which always makes things sound clearer to me. Is that true for everyone?
Pretty much. Increasing compression increases distortion, so we are always playing the distortion/audibility balancing gain. Ironically, users with better hearing can often tolerate more compression even though more severe hearing losses would seem to need it more. Compression affects temporal cues, and with more severe losses pitch cues might be so distorted that temporal cues are all that can be relied upon.
I am a fan, although my understanding is that the majority of audiologists are fitters. I do a bit of both, depending on the patient.
Thanks for the further info on your implementation, @Don. I think to reach anywhere near your level of knowledge, I need to educate myself more and I’ve ordered a used copy of Harvey Dillon’s Hearing Aids book (2nd edition). He probably is not the one to teach me much about DSL (!) but at least with that as a guide, hopefully I’ll begin to understand more of what I’m doing, and less fumbling around in the dark.
Good to hear you’re a DSL fan, too, @Neville. That will help motivate me even more to give DSL5 a try. Right now I’m pretty happy with my implementation, though. Don’t know if you’ve ever listened to Melyvn Bragg’s BBC 4 In Our Time programmes, though. With me being an American English speaker, it always seems that British English is spoken with more sibilant and fricative emphasis (would be interesting to hear what Brits think about this American opinion!). With ReSound’s + fit and perhaps the reproduction quality of the BBC programs, I was always having trouble with British accents lost in background noise such as when I am on a treadmill at the gym. Tonight on a treadmill with my NAL-NL2 fit, all of Melvyn’s British guests came through loud and clear. Maybe it’s just some sort of placebo effect but it does seem to me that my change in fit has definitely helped me with high-pitched consonant sounds without overwhelming me with loud sounds. I will see what I can do with Don’s recipe, though.
Yes, MDB has made several good suggestions on how to improve one’s understanding, e.g. taking relevant courses for free on Audiology Online, the Hearing Aid book by Dillon, and Introduction to Audiology by Martin. I don’t think that I could ever reach the level of understanding of a provider or the more knowledgeable folks on the forum but at least I will be less ignorant.
Too bad the Dillon book has not been updated with a new edition. I asked Thieme Medical Publishing if there were any plans on the horizon to update and a representative said, unfortunately, no. Last published editions of good books remain unfortunately pricey, even as used copies. The decade-earlier first edition of Dillon’s book (~2001) can be had for only $4 as a used book, but not the 2nd edition!
(You’re obviously more knowledgeable on the technical end than I am)