Still trying to determine ideal setting for acoustic and electric guitar playing

[fbacher1]

I’ve been doing some DIY for my Philip 9030’s for my guitar playing. With any of the std settings the shrillness of first 3 strings (G, B E, thinnest) is so annoying I hated to hear myself play and of course this also affects how you play. I’m found that the music program with some reductions can be somewhat acceptable but still not ideal. I’ve turned down 3-8 kHz in the music program by as much as 5-7db, turned off any directionality and noise reduction and have one program with zero feedback control. I also still have to lower the volume in the app by 2.5 to 5 db and by then I start to lose the clarity of my guitar to a certain extent.

I read from a notable musician that boosting 0-500 hz by 4 dB can help a lot but with open domes I can only boost frequencies 1 khz or more. In fact there are zero gains below 1 khz allowed. Changing to closed domes allows changes in the lower frequencies but I am already fighting my own voice and that occlusal effect. so close domes or molds are probably out of the question. Turning my aids off or taken them mutes everything of course and the clarity of the guitar is totally lost.

I wish there was a multiband freq adjustment in the app so I could hone in on the what’s causing all this annoying tinniness with my guitar quickly. I’m not sure but I was thinking of recording my guitar as an mp3, playing back in my iphone and using an equalizer I downloaded for streaming music to find the real offending frequencies rather than this 3-8 khz reduction approach. Pretty frustrated with these 9030’s and one of these days when I have some extra $$$ I’ll just go buy widex aids because I’ve heard for music they are the best out there. Anyone with thoughts on this. I guess I could just reduce a narrower band in step of say, 1-2k, 2-3k, 3-4k, 4-5k, 5-6k, 6-7k and 7-8k by maybe 8 dB to see what works and then identify where the real issue lies. This is further complicated by what sounds good in my music room at home still sounds thin and shrill when I play in a bigger room with my group.

[Noah0]

Hi there, sorry I don’t have any specific answers to your situation but I’m a musician as well and looking to explore similar features with my Philip 9030’s. I have purchased a Noahlink Wireless and wanted to see if you have any links or tips on how I should get started connecting my HAs to my own PC and HearSuite software. Any details on how you got started would be greatly appreciate it.

Thanks in advance!

[fbacher1]

Hi, guitar player here and I’ll tell what is working for me right now. It’s probably not opmtimal but better. I am using two hi fi programs. One with the feedback control turned off. Doesn’t seem to be a feedback issue. I’ve also lowered the freq range from 4-8khz by about 5 dB. then in the app program I do lower the overall volume two steps which is another 5 dB. 1-3 steps might work so experiment. All the noise reduction is off except for soft and wind noise which I also turned off. I think there might be a smaller freq range than 4-8 khz that might be the issue and need to do a lot more experimenting. I did try 2-8 khz too but it was about the same result. I do have the NOAH box so I can make adjustments on my own. If you don’t you need your audi to do it for you. You can only make the app volume changes without the NOAH box and fitting program. The goal is to take the distortion and tinniness out of the first 3 smaller strings while not muffling those tones. Good luck

[Markismus]

Analogue to your problems, I had some keys that simply hurt when playing the acoustic piano. I was able to listen to a beautiful piano after reprogramming them. It worked for Bernafon Chronos 90’s, Phonak Bolero B90’s and Phonak Marvel B90’s. I used the serial Hi-Pro for programming the first two aids and the Noahlink Wirless for the last one.

Following Chasin’s advice (Google it! We also spent quite some threads on it here on the forum. ) I lowered the G50-band (The lowest dB-band) by 6dB (Because the difference in music between the loudest and softest sounds is 18 dB rather than 12 dB as used for speech.)
Then I set the C.R. constant for all frequencies: 1.3 is recommended, however, I’ve even found 1.5 to be quite good. (Compression ratio, the difference between the amplification of the higher bands to the lowest band. E.g. If the G50 has 30dB amplification and the highest 20dB, than the louder the sounds get the more it is compressed. IIRC, the C.R. is calculated as 30/20 = 1.5. ) Now the higher harmonics get the same relative amplification as the base and the shrillness is reduced.

Of course, I disabled all smart options to mangle the sound for better speech understanding. Disabled wind, noise reduction, etc.

However, I really would play via streaming with the electric guitar. It’s useless to go from electric to sound to electric to digital to electric to sound. You could even play via telecoil, it’s mono, anyway.

About those open domes. I remember that in the old days a few people experimented with blocking the sound from the environment with the eraser from a pencil. Just make a small incision, clip it on the wire and shove it behind the dome. It was a real improvement. Nowadays, I would either change the dome whilst playing or stuff some two-component additive silicone behind it. You can get it either at a crafts store or somewhat more expensive as a mold-to-fit hearing protection for swimming or sleeping or custom fit ear-molds for ear-buds.

[fbacher1]

I follow you on the “G50” band but not the compression ratio. I don’t see any place to set or modify compression in the Philips HearSuite program and I already have disable feedback and ALL noise reduction. It works pretty good right now but I think there might be more to gain and will try the 50G change without all my freq mods first to evaluate it. But CR settings are nowhere to be found.for the philips 9030.

Help says: General settings include the features SoundMap Amplification and AdaptiveCompress. You can access General settings from the navigation bar in the Feature Selection section. But nothing about adaptive compression is found there.

It also says this in help: AdaptiveCompress

Amplification will be optimized for different listening environments with AdaptiveCompress. Two controls allow fine tuning if required:

• Compression Control: Speech in Noise Environment controls the amplification balance between audibility of speech and comfort, for speech in noise situations.
• Gain Control: Noise Only Environment will reduce the programmed amplification if no speech is detected. This control determines the amount of attenuation.

The default settings are program specific and can be adjusted individually.

But like I said nothing like that is found in the program that I can see.

[fbacher1]

Oh and I thought about using some foam ear plugs over my domes to 1) keep the low freq from escaping and 2) to block some of the ambient coming in which is 6-8 msec ahead of what the HA is feeding my ear…sort simulatin the benefit with the Widex with it’s 0.5 msec delay.

[Markismus]

@fbacher1

I don’t see any place to set or modify compression in the Philips HearSuite program

That’s what I meant by manually. You have to set the dB amplifications of the G80-band for each frequency. It take me about a minute for each side.
In Phonak Target the C.R. is printed in a row below the amplification table. Also I can switch the visualization of the table, the curves, to C.R.
In Bernafon’s Oasis it was different, but the information was presented.
If that isn’t the case in your brand’s software, you have to choose a constant C.R., calculate the values for each frequency from G50/C.R., iirc, and enter them into the table.

[Markismus]

So here is a list of Chasin’s publication on musiciansclinic.com. I always liked Programming hearing aids for music.pdf (2.2 MB), but I can’t find it in the list.

Here is a screenshot of my program that I use for the piano. It shows both the CR as the bottom row and as a graph for each ear.

My normal routine after making custom earmolds is to do

1. A feedback&real ear test
2. AudiogramDirect
3. Reset the piano music program to a standard music program and disable all options.
4. Reduce G50 by 6-8 dB. (Depends whether the audiogramdirect measurement was just on the threshold of hearing or clear. With 5dB steps the threshold not always ends up on the same loudness. My criterion is that if I can’t differentiate the tone from the noise, it’s below my threshold. )
5. Uncouple the hearing aids, so that I can tweak each side independently.
6. Tweak the G80 values until the curve of the compression ration is nearly straight at 1.5. (I find that selecting multiple sequential frequencies that are too compressed at the same time and increasing them gives a smoother curve than tweaking each frequency individually. However, Bernafon’s Oasis didn’t have that control problem. Also if a bulge in the curve won’t relent, I find that tweaking the G65 value sometimes helps.)
   So I’ve switched the curves around in the software, found that the amplification values in the table switched, too and haven’t been able to get a clear relation. If your software doesn’t calculate the C.R. for you, I wouldn’t be able to program the aids for music. Can’t believe that is the case, though.

This link is geared towards hearing aid compression. Still doesn’t help me to compute the C.R from the values in the table, though. Hearing aids use WDRC (Wide dynamic range compression). So the ratio doesn’t seem to be computed that easily.

[fbacher1]

I’ll ask around on the DIY site about compression ratio and the Philips 9030’s. I looked again last night and can’t seem to get to what the help file says is there. In the middle of a crisis with a cat that had surgury and living in a room away from the other 3 in issolation until at least Saturday. One year old so very active and ready to go, lol.

[fbacher1]

OK, if I lower g50 by 6 db then my gain @ 1K is only 1 dB and that also makes my compression ratio very small like 0.16 @ 1K before I change my G85 value. I am starting with hifi music program that’s a bit different form my general program as well. And the CR goes from 1.17 @ 1K to 3.2 @ 4K and then heads back down to 1.89 @ 8K.

So I should try to adjust all the G85 values to get a CR of 1.5? What about the G65 values? Without any changes are they going to be OK or should they be tweaked as well?

If I go ahead and try to adjust for cr=1.5 after lowering g50 by 6 dB then I’m lowering all the gains from 1K to 2K by 2-5 dB along with the 8K value by -2dB. 3k to 6K either stay the same or are raised by 1-2 dB.

Just looked at your chart again and I guess I got CR wrong because I thought you said it’s G50/B85 for any given frequency and although your chart says the CR is 1.5 that math doesn’t add up. I’m confused.

[Markismus]

That’s what I meant in my last post. The calculation of the CR apparently isn’t straight forward. Luckily your software does it for you.
Qualitatively, the decrease in amplification of louder sounds is the compression. The C.R. would be a parameter that you can adjust to change the compression. This looks like linear compression to me. However, in modern aids non-lineair is used, too. For instance, when the sound level output reaches the maximum power output (MPO) it is not clipped, but compression increases quite dramatically the closer you get to the limit. So how do you quantify such behavior with only one C.R. parameter?
Even worse, we have three points at which the amplification is tweaked 50, 65 and 80. So then there are at least 2 ratios in play.
Looking at the Input/Output curve of the general program for 1kHz:

I see a curve starting at 50, bending at 65 and going to 80. There would be 2 compression ratios. However, there would also be a ratio between the slopes of those two curves.
Coordinates on first curves are: 26,49 and 36,59. So the slope is (26-36)/(59-49)=10/10 = 1. Coordinates on the second curves are: 60,68 and 120,76. So the slope is 60/8=15/4=3.75. De ratio between 3.75/1 is > 3 for 1kHz. Seems straightforward.

Another curve of the general program for 250Hz:

Same kind of curve. Coordinates on first curves are: 28,45 and 46,63. So the first slope is (28-46)/(45-63)=18/8=2.25. Coordinates on the second curves are: 64,71 and 120,80. So the 2nd slope is (64-120)/(71-80)=56/9=6 2/9=6.22. 6.22/2.22= 2.8.

So no success!! The given CR is not the ratio between the slopes.
I am giving up here. I’ve learned that I might me able to tweak numbers. However, I find that I have little idea what the numbers mean:

• if the table changes whenever I change the visualization of the curve.
• if the C.R. is not a ratio between given numbers.
• if the curves are only approximately related to the numbers in the tables below.
• if I look at the introduction audiology textbooks and find that no information is given about quantitative calculation.

It seems that the expert systems, c.q. the software has replaced most quantitative understanding for tweaking. I found that to be the same in fluid dynamics and especially multi-physics simulations. I hope it’s only my limitation.

[fbacher1]

In absence of any information about CR I’ll probably try to issolate which frequencies give me the most tinny and shrill sounds from my acoustic guitar. The electric does the same thing but the acoustic is much easier to notice. Forgot I had my HA out yesterday and was wondering why one of my acoustics that I always thought was too thin sounded pretty good, lol. But without the HA that muffling effect takes the sparkle away too. My hammer approach lowering either 2-8k or 4-8k told me that the issue is in the 4-8k range and probably a harmonic of the fundamental notes being played. It starts sounding tinny beginning with the open G string. I’m not sure I hit 85 dB with the acoustic unless I plug it in but the volume easily hits 65 dB. Not sure what to try next or to lower g50 by 6 dB first. Lots of things that I can try that I could never do with a costco fitter though. I did ask on this forum if anyone knew how to see and modify CR with the Philips program.

That first link you gave me had many highly technical articles. One said that a good music program should first start with a soft speech in noise program and then turn off all noise control and feedback. I’ve always started with the hifi music program and it is way too thing and shrill for music to my ears. It boosts lows and highs over the general hearing programs and it seems those high boost are what trouble me.

[fbacher1]

I did a bit of structural analysis as an ME before I retired and most engineers never really know the math behind the program or even had the time to investigate so some just took the expert system at face value. But there were always ways using various simplified versions and hand calcs, etc. to give you more confidence in the model and the answers. To me that was always a necessary step to gaining confidence in the answer. But so much is hidden in most of these expert systems, like CR in this case. And the apps for my HA takes any user ability to much except raise or lower the volume away so A-B tweaking has to be done in a very slow way by the fitter. Just getting an appointment for one tweak can take a week or more!

[Markismus]

Just checked with my audiologist: The input/output-calculation would have worked if the compression was set to linear. Obviously, since I have little kids, that is not that good a setting for me.

Setting the compression to linear in Target’s global tuning tab. (So not choosing semi-linear and prescribed.)

I get coordinates 30,45 and 40,55 before the bend and 60,73 and 80,87 after the bend. Slope before is 1, slope after is 14/20=0.7. 1/0.7 gives a ratio of 1.4. If I take 30, 45 and 80, 87, I get 50/42 which would be 1.2.
So I can’t generate the 1.3 value from SPL2cc

Getting the Input/output for SPL real ear:

Coordinate before bend: 30,37 and 40,47 is a slope of 1.
Coordinate after bend: 60, 64 and 70,72 is a slope of 10/8 is 1.25 is 1.3

Checking for 1kHz:

Coordinates before bend: 30,59 and 40, 69 results in slope 1.
Coordinates after bend: 60,87 and 70,95 results in slope 10/8 is also 1.25.
However, the number given is 1.2
So I am giving up until I get better documentation.

[fbacher1]

Printed out the values in the general program, hifi music and listening in noise to compare. It’s all so very confusing really as to what to look for and where to start. I can see that music has more gain in the low and high freq. over the general program and that is heard by my ears as extremely tinny and shrill. Totally unacceptable. I found lowering 4-8 khz by 5 db plus the overall volume in the app by 5 dB took care of that while preserving some of the acoustic guitars sparkle. Maybe not all of it and I’m going to just try to narrow that band down in an effort to see if the problem is all of that range or just some of it. I’m never going be an audiologist and just want to get on with playing my guitar and enjoying what improvement in my hearing these Philips 9030’s afford me. I would love to try out the Widex and will do that someday soon.

I’m also going to start with a speech in noise program and turn off all noise control and feedback to see how that compares and works. Baby steps for sure but maybe I’ll learn something without getting too technical and frustrated with this. I doubt my HA fitter at Costco would be of any help. She immediately goes to the fine tuning help program in HearSuite to look for advice and she has 35 years experience and wears HA herself. She is just not the right person to help me unfortunately.

[e1405]

Hi! Here is my experience crafting a music program for guitars, maybe something in these two threads could help you (and I’d love any feedback to improve what I’ve done so far). Cheers!

Tweaking music program step by step

No matter what my audi does, my acoustic and classical guitars always sound awful when I get home after a session with her. I have done some reading (references 1, 2, 3, 4, 5) and devised a method to improve the musical experience. As a layman in the field, I would like to hear your comments and suggestions on things I got wrong and ways to improve this little experiment: Spend some time experimenting with domes and molds. Find the perfect acoustic setup for you. My ear canals are too narrow …

and

DIY works - my journey has paid off (and how I did it)

I have been self-programming my HAs for the past 5 years. After hundreds of hours and many frustrating programs along the way, I think I nailed it! This is me listening to a podcast in a quiet room today: [signal-2023-05-18-214334] The sound of the podcast is so clear and crisp, and I comprehend everything with either ear at 1/4 of full volume. For so long, my default was 3/4+ of full volume. This is my main program against the target gain suggested by Genie/NAL-NL2. [P1 vs2] My music pro…

[fbacher1]

I’ve done some of this already but any suggestions about setting the compression ratio to this or that doesn’t work if you can’t see the compression ratio and the software doesn’t let you adjust it like with Philips. That’s my issue right now. So I’m just stabbing a multiple changes to one of the 3 sound level and testing hoping something works and that involves an infinite number of combination with Philips. I need some Philips 9030 specific advice at this point to get much further with this. I’ve also found certain voices to be tinny at time and have issues with my own voice as I increase the gain (I’m at 90% of the target right now and was at 80%). I’m ignoring the basic hearing part right now to focus on guitar and have something that “works” and maybe it’s good enough and there’s no more to gain. Maybe a little bit muted compared to the ideal but much better than shrill and tinny by a long shot. Thanks

[e1405]

@fbacher1

I’ve done some of this already but any suggestions about setting the compression ratio to this or that doesn’t work if you can’t see the compression ratio and the software doesn’t let you adjust it like with Philips.

Have you tried to click CR?

[fbacher1]

DAHHHHHHH, why of course not! Can’t see the forest for the trees. Thank you so much…maybe need to get my eyes checked next!

[fbacher1]

So here is what I have to start with in hifi music program. This is even more treble and very shrill for me compared to even the general program so first step is to reduce the G60 by -6 dB?

That seems to lead to having to increase the level of the others a lot if I’m trying to hit 1.3 CR so not sure about this yet. I’m just guess but I don’t think I ever get into the G80 levels with my music or if so just barely. so maybe just leave that one alone for now?