Notched Music Therapy (DIY, AudioNotch, etc.)

[James White]

I'm working on a tutorial, but if anyone wants to get it working before, here is what you'll need:

Free open source tool :
http://sourceforge.net/projects/equalizerapo/

This will work on it's own but it's tricky to use. There's a documentation in the wiki section.

Paid software that will work :
http://ntonyx.com/vac.htm $50 (this is exactly the same as Soundflower) + with any DAW software like Cubase, Ableton live, etc (but all these a pricy if you stay "legal" hehe)

 

[spleen]

Well, the software's I use are :

• Any music player like VLC or foobar2000 where you can specify the output to the device you want
• Virtual Audio Cable, it creates a virtual sound card so you can redirect any sound flow to his input and capture the output
• Ableton Live, the notch is done with this powerful software in real time, and you have nice graphics to see what's happening so it is very useful to adjust everything

Here the tutorial :

First, when you install Virtual Audio Cable you should verify it creates a default "cable" aka virtual sound card, else create one.

​

​

Lunch your music player (foobar2000 in my case), and go in preferences/playback/output and choose your virtual sound card (Line 1 Virtual Audio Cable in my case). Start any music, you should not hear anything.

​

Lunch the Audio Repeater tool of Virtual Audio Cable and select your virtual card sound (Line 1) in Wave in and None in Wave out. Click Start, you should see the bottom blue bars moving.

​

Lunch Ableton, go to preferences and select your virtual sound card (Line 1) in Audio Input Device. Verify you have your speakers selected in your Audio Output Device.

​

​

Then, in the main screen, click on the small recording button. You should hear the music you lunched with foobar.

​

​

If you hear the music, the workflow is done, we have now to modify the signal with Ableton. We will do a notch in this case. So, go to Audio Effects and double click on EQ Eight, you should see a small window appear on the bottom. Unselect the small orange boxes minus the first one and click on the notch icon.

​

Adjust the notch frequency as you want; your music is now notched in real time. Notice that you can adjust the notch bandwidth also, add several notched points, high or low pass filters… if you know what you want, you can do it with Ableton.

​

​

Now, we will add a spectrogram filter to see the modification of the signal in real time. Go to Audio Effects and double click on Spectrogram. On the Spectrogram window, click and the small black arrow.

​

We can clearly see where the notch occurs. If you unselect the orange box of the notch filter on the EQ Eight filter, you will see the signal back to normal.

​

​

For a better filter quality but more CPU consuming, you can right click on the EQ Eight filter and select Hi-Quality; and for a finer spectrogram, select a higher block size.​

Notice that I wanted specifically to be able to control which sound flow I will modified, only the one from the music player, not all the sound flows from my computer, so when the workflow is lunched, I can use my computer and hear the others sounds from my computer normally. If you want to modify the sound flow from spotify or an internet radio, you have to configure the input workflow differently, but the principle is the same. Regarding Ableton, a version exists for OS X and also for Linux (in combination with Wine) and creating virtual sound cards is quite easy (and free) on these systems. Anyway, if you understand the concept of this worflow, you can do the same with other softwares and other systems.​

The good thing about this workflow or the other one made previously for OS X is that it should be in theory possible to transpose them on tablets (on steroids) with similar systems. The new generation of small tablets will allow us to have kind of a portable music player with our home made real time equalizer. No more scam from the expensive pseudo-audio therapies showing around which are not tunable at all unless you come back to see the doc…

 

[James White]

Thanks for the tutorial Spleen.

For everyone trying this technique, it is very important to inspect the sounds you are listening to. Some mp3 files have a highfrequency cutoff so the notch would be ineffective.

To analyse the sound that is processed, you must use a spectrum analyzer. If you have a high frequency tinnitus, you can also slightly compress the frequencies above your tinnitus so that they emphasize the notch (most of the higher frequencies are usually dimmed).

 

[spleen]

James is perfectly right. Take care that the spectrum of the music files you want to use includes your tinnitus pitch and go further, else the notch will be useless. Most of the time, music compression cuts the high frequencies, so it's better to use high quality compression format for MP3 like 320 CBR or better, a lossless compression format like the FLAC. You have the Apple equivalent for all of these formats if you use an IPod. In my case I always start from the FLAC format files (you can buy them online, convert a CD to this format or "download" files already in FLAC…) and convert them to MP3 320 CBR to ensure I have the best quality/compression possible, not loosing high frequencies.

 

[James White]

I've made a video for this on the mac

 

[Esko]

I found a good guide using AU Lab from Apple: http://best-mac-tips.com/2013/01/31/free-system-wide-audio-equalizer-mac/

I saved my presets made in Garageband and opened them in AU Lab. The benefit of AU Lab being not having to run Garageband which consumes more resources. Garageband is not that much at 125 MB, but worth noting. Not sure if the Garageband track can be started at boot like AU Lab can.

To select a more precise frequency, hold Option. I would like to enter an exact one, but haven't found a way to do that.

In case you can't access AU Lab from Apple, I uploaded the files. AU Lab.app.zip is taken from xcode44[...].dmg and zipped. au_lab.dmg appears to be an older version.

 

[Job]

I'm not sure if this is in another thread, but ...

My tinnitus is sometimes easier to mask if I play white noise and music at the same time ... does anyone know of a way to combine 2 tracks on a mac? Feel free to just conversation me as I do not want to clutter this thread with different topics. Thanks for any help.

 

[Esko]

Remember to disable the notching by selecting Headphones again instead of Soundflower in System Preferences -> Sound -> Output if you listen to already notched audio or use a program like Tinnitus Tamer.

 

[jazz]

@James White Thanks! Very easy!

 

[James White]

I found a good guide using AU Lab from Apple: http://best-mac-tips.com/2013/01/31/free-system-wide-audio-equalizer-mac/

I saved my presets made in Garageband and opened them in AU Lab. The benefit of AU Lab being not having to run Garageband which consumes more resources. Garageband is not that much at 125 MB, but worth noting. Not sure if the Garageband track can be started at boot like AU Lab can.

To select a more precise frequency, hold Option. I would like to enter an exact one, but haven't found a way to do that.

In case you can't access AU Lab from Apple, I uploaded the files. First link is taken from xcode44[...].dmg and zipped. au_lab.dmg appears to be an older version.

AU Lab.app.zip (5.8 MB)
xcode44audiotools6938117a.dmg (34.6 MB)
au_lab.dmg (7.3 MB)

This is brilliant Esko ! Plus, I guess AU Lab is not a "stupid" ReWire host. I spent hours trying to disable ReWire on Garageband because i use Ableton Live to make music and you can't launch both app simultaneously ...

 

[Esko]

After rebooting I found that AU Lab did not load my notching track, which resulted in no sound at all since I have Soundflower as output in System Preferences -> Audio.

This should be a solution to that:

 

[Arkadiusz Goralski]

Hello!

This is my first post, so at the start I'd like to say hi to the community!

Since I'm a mac user too, I'd like to thank James and Esko for the tips on how to set up the software to noth the audio system wide. But I have a tip I'd like to share, thus my post - you can actually set the frequency by typing it in in the plugin settings - just change the view to generic.

 

[Esko]

Hello!

This is my first post, so at the start I'd like to say hi to the community!

Since I'm a mac user too, I'd like to thank James and Esko for the tips on how to set up the software to noth the audio system wide. But I have a tip I'd like to share, thus my post - you can actually set the frequency by typing it in in the plugin settings - just change the view to generic.

Welcome, and thank you! I was looking for that but didn't find it.

 

[Per]

Found it close to impossible to match my T frequency.

 

[Per]

I tried Audionotch and the site is well designed, user friendly with step by step instructions. My problem is that I had a hard time pinpointing the frequency of my fluctuating T and it seems to be over 12K so this does not really benefit me.

Same here. Mine seems to fire at several frequencies at the same time. Don't know if this is just an auditory perception but it fluctuates, making it very hard to identify the exact frequency. I hope an audiologist can help me to identify the frequency(s) Mine is also very high pitched, although I don't know how pitched cause I have no other T to compare it with.

 

[Steve]

Same here. Mine seems to fire at several frequencies at the same time. Don't know if this is just an auditory perception but it fluctuates, making it very hard to identify the exact frequency. I hope an audiologist can help me to identify the frequency(s) Mine is also very high pitched, although I don't know how pitched cause I have no other T to compare it with.

I put a really basic video on YouTube that may help. It moves in 1000Hz steps, it may help you to identify a band where your T lies.

 

[Peter Phua]

Found it close to impossible to match my T frequency.

Matching is really hard. If you do a pubmed search on "tinnitus pitch matching" you'll see that it's basically unreliable for about 50% of people.

 

[Peter Phua]

Found it close to impossible to match my T frequency.

I wrote a blog post summarizing one of the papers discussing the difficulty some people have with matching their tinnitus pitch frequency:
http://www.audionotch.com/blog/tinnitus/more-on-tinnitus-pitch-matching-a-study-of-interest/

Disclosure: I run a Notched Music software company.

 

[Owch]

Reverse engineering the Neuromonics 'Sanctuary'?

I just had a look at what is offered by the Neuromonics 'Sancturary' device while discussing hearing aid companies in another thread and I am distinctly underwhelmed:
https://www.tinnitustalk.com/thread...earing-related-health-issues.2128/#post-18841

It appears to be a customised, lossless audio file player which offers the user the choice of playing one of three pieces of music. Each piece of music appears to have had a signal mixed in that corresponds to their measured tinnitus frequencies. It may also have had some equalisation applied which corresponds to their audiogram (this second point is not certain but is one interpretation of some of the device's marketing phrases).

Does anybody know somebody that has a 'Neuromonics Sanctuary' device? It would be very interesting if they were able to study the files on this device (or simply the recorded output) to see what signal has been applied and how it corresponds to the measured tinnitus frequencies. It would also be interesting to see if there has been equalisation applied and how that corresponds to the patient's measured audiograms.

Obviously Neuromonics has copyrites on their product and distributing music files recorded from the device would probably be illegal, but I am interested in knowing what they did to their files (ie. are they just notched?) - and the equalisation aspect is particularly interesting as this doesn't seem to have been covered so far in this topic.

 

[JohnJ]

I've had bad tinnitus since mid August and I've just found this forum. I've been playing around with notched music the last couple of days and it seems to me it's a sinch to produce on a Windows PC using Audacity (free and open source) as it has a notch filter built in. You just open your music file and then under the Analyze heading you select "Plot Spectrum" to check your music includes the frequency of your tinnitus. Then Under "Effects" you select the Notch Filter and enter the frequency of your tinnitus. There is another variable "Q" that you can set which adjusts the width of the notch. If I'm correct the default value of "1" equates to 1 octave and higher values make the notch narrower. I've read on here and elsewhere I think, that notched white noise appears to be more effective than notched music. This makes sense as I would think white noise includes equal amounts of all frequencies whereas music might only have a small fraction in the required region - especially if your tinnitus is quite high, say 8khz and above. With this in mind tonight I'm mixing some white noise with some music and then notching the mix. Hopefully it will be easier to listen to than just white noise and just as effective.

Anyone else played around with notching in Audacity?

John

 

[Neenie]

What do you guys mean when you say you listen to "notched" music. Oh wait, I just googled it and founds this:
The frequency spectrum of the music was modified for each patient individually by means of digitally filtering out the frequency band of one octave width centered at the individual tinnitus frequency (i.e., the frequency that sounds like the tinnitus). The problem is that I can not find my tinnitus frequency. This makes me mad because I feel like it means I can't be treated, if I can't even recognise the pitch then how can I ever make it go away?

 

[Neenie]

Watch this You Tube video.
It teaches you how to notch your own music. I've just done it, not that hard to follow. This is an article describing how it works: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2918775/
You may as well try this. It can't harm!

 

[Markku]

It's not unusual to find matching one's tinnitus frequency difficult. Even I do even though mine's quite constant in tone.

 

[Linds]

What do you guys mean when you say you listen to "notched" music. Oh wait, I just googled it and founds this:
The frequency spectrum of the music was modified for each patient individually by means of digitally filtering out the frequency band of one octave width centered at the individual tinnitus frequency (i.e., the frequency that sounds like the tinnitus). The problem is that I can not find my tinnitus frequency. This makes me mad because I feel like it means I can't be treated, if I can't even recognise the pitch then how can I ever make it go away?

How did you find your T frequency Neenie?

 

[Markku]

How did you find your T frequency Neenie?

You could use this: http://www.audionotch.com/app/tune/
Or this:

 

[Linds]

Thanks Markku. According to this audionotch I can hear up till 11khz in both ears and 12khz in right nothing after that.
Does that mean I have hearing damage at those frequencies and or that is my T sound frequency?
Thanks for your reply.
Im going to an ENT in Feb will they do this hearing test?

 

[Markku]

No. The idea is that you try to match your tinnitus sound with the tone generator's output, and once you've found a match, you'll see what your tinnitus frequency is (the Hz field and the number next to it).

Im going to an ENT in Feb will they do this hearing test?

Usually they test your hearing up to 8 kHz, because those cover the frequencies of speech. It's possible to test the higher frequencies, too, but you likely need to specifically ask for that.

 

[Neenie]

Hey! I used Tinnitus Tamer to work out my frequency, see here: http://www.vavsoft.com/Tinnitus_Tamer.html
Do it over several days or several times to work out which frequency you have. I found this insanely difficult, and I am still unsure if I got it right. Butreally, I don't think you have to be too specific. You then get some relaxing music you like, then notch it by getting your tinnitus frequency, then using Cool Edit, see here: http://cool-edit-pro.soft32.com/fre...396&pl=&ds=s&gclid=CNCz1_L-wLsCFccTpAodmgcAQw you dampen out the one octave around your tinnitus frequency. Then listen to it for 2hrs every day and hopefully you will feel better after 6 months. Yer, I know. It's ages, but we have little choice in it really. In Australia if you get a hearing test they test you over 10kHz too so that's good because if you have high frequency hearing loss it will be detected.

 

[Neenie]

Hi all. I just subscribed to Audionotch, and notched some waterfall sounds. I'm listening to it now. I can very clearly hear my tinnitus with this as compared to regular waterfall/white noise sounds. Does this mean I have the tinnitus frequency right, because the tinnitus frequency is being "notched out" therefore I will hear it clearly on the mp3 file? Feeling more anxious listening to the notched sounds compared to normal sounds. Is this normal?

 

[Sparks60]

Neene, have you took a look at this site yet to see if it might help you,
http://www.generalfuzz.net/acrn/
i have been using to it for about 1 month now and has lowered my T down.
mine was as bad as yours for many years ,30 or more so i feel your pain.
i have also found that 1 Niacin 500mg and 1 thiamin 250mg(B1) taken daily(for about 3 weeks now) has really lowered my T to where i can sleep at night. at the moment i would have to say 80% lower, (its still there but i have to listen to hear it now, how wounderfull this is)i told my wife today that this past week has been best ever with tinnitus, a tear ran down my face for joy. I just wanted to share this info so you can see there is hope.