Acoustic CR® Neuromodulation: Do It Yourself Guide

Here are figures based on spectrum analysis in Audacity of the capture from the device compared with frequencies stated in your webpage:

-----ANM: 6140 7196 8808 11217 -------- Tinnitus: 8kHz
Gen Fuzz: 6139 7202 8772 11186

Listening to them I at first thought that your tone sequence moved slower than the device, but comparison in Audacity shows any difference to be tiny (if present at all). There is a slight ticking noise apparent in the tone sequence produced by the webpage though - clearer at higher volume than the tones should be used at. I wonder if this might have any impact on efficacy or safety considering that the sequence might be listened to for up to 6 hours a day. Also some random tones seem to creep in occasionally, although I doubt they would be a problem as they are so intermittent.

I did not check the order of tones in the sequences - if I remember my reading of the papers correctly then they should be random so it does not matter?

Awesome achievement!

 

I used to mess around with amateur speaker design and can confirm that higher frequencies are more directional then lower ones. Some models of speaker driver will also tend to disperse the higher frequencies wider than others. It is definitely worth using headphones for frequency finding or listening to the tone sequences.

 

Do you hear the ticking noises at the beginning or the end of the tones? I can change the shape of the envelope to ensure they don't occur.

The pattern in the sequences is from the paper - random within some parameters.

 

Hi,

I can't be sure if the ticks are at the start or the end, but I reckon you'll hear them if you turn up the volume on the page and the computer master volume. That is, unless the problem is really my headphones being on the fritz from all that turning up the volume.

The ticking noises probably aren't a problem, as they are not perceptible at the volumes at which the tones should be listened to. But if it is easy to play with the envelope until they vanish then p'raps that is worth doing.

 

With apologies for hogging the thread, there is another beneficial thing that they do on the trial - following the ANM protocol. I’ll try to describe that here with a suggestion to emulate the effects of this using the General Fuzz page. Apologies for complicating things, but this might potentially improve the effectiveness of the DIY approach.

Once they have identified your tinnitus frequency (or the one to be treated if you have several), they will know the four tones which the device will play. They then measure the threshold at which you begin to perceive each of these tones. This gives them an idea how loud each tone should be played through the device. They then present the tones to you in pairs and you adjust the volume of each until you perceive them all as equally loud.

The crucial point is that for most people (especially people with hearing damage) these tones will not be equally loud in an absolute sense. But the goal is that they are equally loud for you – i.e. the stimulation received by your auditory cortex is equal for each tone.

Here is a comparison of the amplitude of my tones compared with those from General Fuzz’s webpage. Mine are of clearly varying amplitude, which reflects the profile of my damaged hearing and its different sensitivity to different frequencies.

 

If you hear the volume of all frequencies played by General Fuzz’s ACRN webpage as being pretty much equal then you can happily ignore what follows.

If you don’t and are serious about using this resource to try to treat your tinnitus then it is my suggestion to modify the amplitude of the tones so that you perceive all as of equal volume. One possible way of doing this might be as follows. Ideally you would use the same headphones or earphones that you intend to use to listen to the tones on your MP3 player or whatever other device. This then compensates for the unequal frequency reproduction of those earphones as well as your ears!

In Audacity generate a series of four tones matching the four frequencies specified by the webpage.

1- Use GENERATE --> TONE and choose a sine wave of the first frequency, amplitude 0.2, duration 0.5s.
2- Click the button just to the left of RECORD (red circle) to go the end of that tone. This is a crucial step, otherwise things get messy.
3- Use GENERATE --> TONE and change amplitude to 0 to create 0.5s of silence.
4- Click the button just to the left of RECORD (red circle) to go the end of that silence.
5- Repeat this sequence for each of the four tones. You should then have a sequence that plays each of the four tones in turn.

6- Click EFFECT --> EQUALISATION.
7- Adjust the equaliser and press PREVIEW until each of the tones sounds exactly the same volume to your ears. Then click OK.
8- Now import the audio file you made by recording the tones generated by General Fuzz’s webpage (or record it now).
9- Apply the same equalisation to these tones. Export the file again and it should be ready to use.

You should only have to do steps 1-5 once, then if you don’t change the equaliser settings then you can just reuse as often as you retune your tones based on the movement of your tinnitus frequency.

I hope this is useful – it is only a suggestion to try to help everyone get the most from this excellent resource.

A final suggestion – be cautious. You can make your tinnitus worse (hopefully only temporarily) so don’t try it if you are feeling desperate or likely to react badly to a spike. And don’t play the tones too loud – they only have to be just audible.

 

ANM Adaptive Neuromodulation GmBH filed for bankruptcy with the Cologne Court in Germany on 1 October 2013.

See link with the filing:

http://www.indat.info/insolvenzen_gerichte/K%F6ln.html

 

Thanks for the link. I wonder how they will continue to support the devices already sold, and who will take over the sales and equipment? Seeing as they have invested in a research project here in the UK I imagine somebody will carry on providing it.

 

I've just tried the generalfuzz page and trying to match my T frequency was a nightmare! I'll have to try again later, listening to the high pitched single tones actually seems to have made my T hugely decrease, to the point even with my ears plugged with my fingers I can't hear the tone

Has anyone got any suggestions of an easier was to match my tone than the slider? I could have easily convinced myself that I had about 8 different tones between about 6khz and 8khz (when in actual fact I have one clear frequency which is the same in both ears)!

 

Since my tinnitus frequency is 14kHz (sadly) the upper frequencies of the sequence go pretty high. I am quite hard of hearing up there. Would it be safe to boost those frequencies so I could hear them, no?

 

I updated the acrn web-app @ http://www.generalfuzz.net/acrn/

Changes:

• There is no longer a "clicking" noise at the end of each tone.
• The text of the frequency tone being played turns red - this gives a visual cue to what tone is being played
• I added the ability to adjust the individual volume for each tone.
• The app will recall the settings from previous usages.

Suggestions and feedback are very welcome.

 

Sarah - matching your own tone can be extremely difficult. I cannot do mine on my own. If there is anyone who can help you by playing the tones while you sit with your eyes closed and say "higher" or "lower" then that might be a little easier. Get that person to make a list in numerical order of the frequencies tested and how you responded so you can review your answers later. Do it several times, don't worry about occasional inconsistencies, and try to hone in on the tinnitus frequency gradually. Practice seems to help a bit.

JTP - I think ANM say (or said!) that the treatment is only for up to 10-12kHz (can't remember exactly). You could try it, but the highest tone in your sequence that you would be listening to would be nearly 20kHz. Are you sure you can actually hear that high, even with a little boosting?

BTW in my post #216 above I wrote complete nonsense when I said "You should only have to do steps 1-5 once...". Unfortunately you would want to do all steps each time you established a new tinnitus frequency to treat. Sorry for being a dope! And in a perfect world we would do this for each ear as the pattern of hearing loss might well be different in each ear. But since we are not treating each ear with different tones, and we also have lives to lead, hopefully it is OK to assess the relative volumes of tones in both ears together. Hopefully...?

 

@generalfuzz - awesome improvements! So much easier to adjust each tone than my longwinded way!

 

Thanks for your hard work generalfuzz! It's not working for me yet, but I sure appreciate that little more hope

 

@JTP as dboy said, the treatment is only supposed to be for T up to 10Khz. If you start to boost the much higher frequencies to match the volumes you will most likely cause hearing damage as they will end up very loud to compensate for the reduced hearing up there.

@dboy I would say you're right, when it sounds about right in both ears you should be fine. Unless you know of hearing damage to one ear, or can really perceive a difference in levels.

@generalfuzz Awesome updates.

 

@dboy , certainly not, I am -70dB at 15kHz. OTOH I soldered a piezo transmitter to TSR plug to make a bone conduction device, I might try that

 

Hi everyone.

This is my first post here on tinnitustalk. I've been following this thread for a week now and just wanted to say thanks to everyone who contributes to this topic.

Initially, before discovering this forum, i stumbled across the "100 - 10khz random" video by Steve and thought i try to improve the envelope of the tone. I managed to do so by using a FM synthesizer in FL studio.
Acoustic CR random tones 10K.mp3 basically is randomly generated tones from 150 - 10Khz. The timings are off but i think the tone is almost identical. I listened to this mostly because i thought/hoped that if anything it will somehow "calibrate my tonal map" so that i can find the matching frequency/frequencies easier. No such luck yet.

In addition to the ratio of the other frequencies i also think that the characteristic of the signal is very important. So far all the samples from this thread have had other spectral components, at the beginning and at the end, probably because simple ADSR envelopes were used. This, i think, makes the impulse less precise and more tiresome to listen to for extended periods of time. The original device generates a very smooth tone.
The FM synthesizer (sytrus) can make a pretty decent, soft, pure tone. However, when i used the method described at the beginning of this thread but by using Adobe Audition instead of Audacity i found that one could make a soft, pure tone with almost no other harmonics and zero attack and release.

I don't know if there's an option in other wave editing software for it, but in Audition you can make curved fade in and out envelopes



And make the signal look like this



I guess all of this is moot because you would have to arrange all of the frequencies manually and @generalfuzz already made an easy to use generator ; which i might add makes the smoothest tones so far (after the method described above).

Just for comparison

Top - general fuzz generator
Bottom - made in Adobe audition



@generalfuzz

Is it possible to:

1. Make the envelope smoother
2. add a WAV file export to your site, preferably 48Khz 24 bit
   
3. add an option to add more base frequencies and stereo separation (for those willing to experiment)

and could you (or somebody else) please explain the beaning of "cycle frequency", is delta the same thing?

I read the thread over on reddit.com and tried the generator from /u/zyl91. Although the algorithm is not in line with what we've since found out here, the program itself is pretty nifty in the sense that you can add multiple frequencies, export WAVs, have stereo separation and change the cycle frequency (whatever that means )
I really do appreciate what you're doing here and hope i'm not asking for too much.

Sorry for the wall of text/images.

Thank you.

 

1. Not really. The library I'm using the generate the tones really is just a pure tone sine wave oscillator. The only way I have to shape the tone is using an ADSR envelope. I spent a good while dialing in that envelope making it as smooth as possible given those restrictions.
2. Yes. This to me is the key missing feature. I have proven to myself that it is indeed possible to do this, but I haven't actually figured out how to overcome all the technical issues yet. The next time I have some free time to work on the app, I will be focusing on implementing this. It's going to be 44.1k 16 bit, but I don't think that will make it any less effective for sound therapy.
3. Probably not for base frequencies. It will make the interface too complex, and I don't see the advantages. I may do something with panning eventually, especially if you want to have different sequences for the left / right ear.

 

The ACRN won't play on my iPad...

Help!

 

Has this ever worked for anybody? Or is it just a theory

 

It's not going to work on a mobile device. It requires a computer with a modern web browser (chrome or firefox most likely)

 

It stopped working for me on Firefox, but works great on chrome.

To repeat Erlend's question, has anyone tried this and had any positive results? If so, what sort of time period did it take to notice any difference?

 

That would interest me too

 

How long should I run this for? 5 mins? 2 hours?
Mine is around 8k, I notice a suppression in the ringing with this on.

So does that mean even if I have no hearing damage and might have TMJ from dental related things when I was younger that this will still work?

 

But when you have a bunch of sounds in a hissing pattern, it's not smart to try?

 

How loud should it be? I have it on 50% headphone volume and 100% volume on the website. I usually listen to music and other things at around 15-25% volume on windows.

 

6500Hz pattern
•4980
•5848
•7137
•9094

8000Hz pattern
•6140
•7196
•8808
•11217

Using these two sets of ANM frequencies, we got the following formulas for Tinnitus ranging from 6500Hz to 8000Hz:
f1=0.7733*T-46.6667
f2=0.8987*T+6.6667
f3=1.1140*T-104
f4=1.4153*T-105.6667

I hope people can provide even more ANM patterns, so that we can calculate the formulas for all range of Tinnitus.

 

Hi,

I got my device readjusted. I don't think it is working at all for me. The new frequencies were 7.5kHz in one ear and 8.5kHz in the other. Here are the patterns recorded from the device and analyzed in Audacity. I hope they are useful.

7.5kHz : 5744, 6752, 8249, 10511
8.5kHz : 6509, 7644, 9346, 11909.

Baddream - Sounds like you are playing the ACRN too loud. You wanna listen for 4-6 hours a day at a volume that is just audible - i.e. very quiet. "Soft and clear" I think is what they say on the trial. You should be able to go about your normal activities without the ACRN interfering in any way. Best thing is probably to stick it on an MP3 player and use earphones that allow environmental sound in unobstructed so you can get on with your life while the ACRN is (hopefully) working.

 

Hi @dboy, thank you very much. Real ANM patterns are the most precious things for us. I will analyze these new data as soon as possible.