• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Beta Test: Earful -- a hearing test App

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
#1
Based on the discussion and a poll earlier, I'm starting a new thread to beta-test Earful. An app designed to measure the lower threshold of audibility for a range of frequencies.

https://distortaudio.org/earful.html

The app is free to use. Some instructions on how to use it are at the above link. And while I took every precaution I could think of, I'd like to caution everyone to be very careful when testing with Earful in order to make sure that loud sounds are not played through your speakers or headphones to avoid possible equipment and hearing damage.

Set the maximum volume appropriately, adjust your preamp volume control, turn off Windows sounds to avoid loud beeps or notifications. Use Exclusive WASAPI mode if your device supports it.

Please post your results here, in a graphical form or as a saved frequency response file from Earful, if you're willing to share them! Have fun!

1595446669618.png
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
5,540
Likes
10,024
Location
The Neverlands
#2
When a used headphone has a 10dB spike or dip or severe roll-off wouldn't this throw the measurements off ?
How do you safeguard actual SPL ?
Are there requirements for headphones or a list with preferred headphones ?

Do you use noise bands or a tone ?
Noise bands could be more accurate.
 

FrantzM

Major Contributor
Forum Donor
Joined
Mar 12, 2016
Messages
1,603
Likes
1,903
#3
Avast is saying the type of application is new and is analyzing it... they'll let me know :(
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #4
When a used headphone has a 10dB spike or dip or severe roll-off wouldn't this throw the measurements off ?
How do you safeguard actual SPL ?
Are there requirements for headphones or a list with preferred headphones ?

Do you use noise bands or a tone ?
Noise bands could be more accurate.
All settings are in dBFS. You can set the maximum allowed dBFS value at the top and Earful will not allow you to adjust above that point. The default maximum is -20dBFS.

I'm planning on adding an EQ function to adjust for headphone response, but so far, I've found my curve to be fairly consistent across multiple headphones I've tried. You can certainly use any existing EQ if you have it available to correct for specific headphones response.

You have a choice of single tone, warble, or white noise.
 
Last edited:
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #5
Avast is saying the type of application is new and is analyzing it... they'll let me know :(
Of course, anything new will look suspicious to them. You may need to tell it to trust that you know what you're doing.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #8
When a used headphone has a 10dB spike or dip or severe roll-off wouldn't this throw the measurements off ?
How do you safeguard actual SPL ?
Are there requirements for headphones or a list with preferred headphones ?

Do you use noise bands or a tone ?
Noise bands could be more accurate.
@solderdude , thinking about providing SPL calibration for headphones, would the 'CD + SPL meter' be a good way to measure a reference level? Would you recommend a single tone or noise (white or pink) to use for measuring the reference level?
 

zermak

Senior Member
Joined
Jun 2, 2019
Messages
315
Likes
203
Location
Italy
#9
I am more interested in my hearing threashold but being able to compare my results with others is a nice plus.

Anyway I gave it a try last night but my noisy tower in the quiet room didn't help... I will pull out my fanless laptop tonight :)

Note about the program. Loading the ISO226-2003 curve messes up the Y-scale and has a weird shape (not like intended) while the Fletcher-Munson doesn't load at all but both files are fine (I checked with a text editor and in REW). Another issue is that it kept playing a tone while I closed the program (the executable was still working in background) and I had some glitches while playing some tones (maybe I was too rushy changing tones and loudness but wanted to let you know).
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
5,540
Likes
10,024
Location
The Neverlands
#10
You have a choice of single tone, warble, or white noise
great !
@solderdude , thinking about providing SPL calibration for headphones, would the 'CD + SPL meter' be a good way to measure a reference level? Would you recommend a single tone or noise (white or pink) to use for measuring the reference level?
Yes, that will work. Can even do it with a phone+amp provided there is a good seal between the phone mic and the hole in the CD as well as CD + earpads.
Perhaps a narrow 1kHz noiseband ? This way A, C or other weighting of the SPL meter doesn't matter.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #11
I am more interested in my hearing threashold but being able to compare my results with others is a nice plus.

Anyway I gave it a try last night but my noisy tower in the quiet room didn't help... I will pull out my fanless laptop tonight :)

Note about the program. Loading the ISO226-2003 curve messes up the Y-scale and has a weird shape (not like intended) while the Fletcher-Munson doesn't load at all but both files are fine (I checked with a text editor and in REW). Another issue is that it kept playing a tone while I closed the program (the executable was still working in background) and I had some glitches while playing some tones (maybe I was too rushy changing tones and loudness but wanted to let you know).
Thanks! Yes, quiet environment is key. Despite having headphones on, I can hear the noise the hard disk makes which is a few yards away, and it interferes with finding the lower audibility threshold :)

Interesting problem with loading external data. You can change the scale by zooming in or out on the left axis (use mouse scroll wheel to do it, after clicking on the axis). Did you change any of the default settings before trying to load the data, and did you try loading it as data set 1 or 2?

I'm suspecting that perhaps your regional Windows settings are affecting it. The data files use a decimal point instead of a comma, maybe that's confusing Earful. Which setting do you use in Windows?
 

zermak

Senior Member
Joined
Jun 2, 2019
Messages
315
Likes
203
Location
Italy
#12
Thanks! Yes, quiet environment is key. Despite having headphones on, I can hear the noise the hard disk makes which is a few yards away, and it interferes with finding the lower audibility threshold :)

Interesting problem with loading external data. You can change the scale by zooming in or out on the left axis (use mouse scroll wheel to do it, after clicking on the axis). Did you change any of the default settings before trying to load the data, and did you try loading it as data set 1 or 2?

I'm suspecting that perhaps your regional Windows settings are affecting it. The data files use a decimal point instead of a comma, maybe that's confusing Earful. Which setting do you use in Windows?
I am using Windows with English language (even if I am Italian) and, I don't know if it matter, using the International keyboard layout.
I have cleared all the settings (with the button provided) and tried to load the curves in both set 1 and set 2 with same issues and actually on set 2 the ISO2003 outputs errors on the graph tab and errata corrige, the Fletcher-Munson curves is like a flat line with a slope starting at 500Hz (and well I used the zooming out option but I didnt go that further and I have now, the curve is at around 750000dB :D).
Anyway I'll try to makes the changes (switch the commas and dots) on the curves text files and report back.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #13
I am using Windows with English language (even if I am Italian) and, I don't know if it matter, using the International keyboard layout.
I have cleared all the settings (with the button provided) and tried to load the curves in both set 1 and set 2 with same issues and actually on set 2 the ISO2003 outputs errors on the graph tab and errata corrige, the Fletcher-Munson curves is like a flat line with a slope starting at 500Hz (and well I used the zooming out option but I didnt go that further and I have now, the curve is at around 750000dB :D).
Anyway I'll try to makes the changes (switch the commas and dots) on the curves text files and report back.
Just tried changing Windows setting to comma for the decimal separator, and clearly reproduced the problem you reported :) I'll add a fix for this, so that you don't need to change the files.
 

zermak

Senior Member
Joined
Jun 2, 2019
Messages
315
Likes
203
Location
Italy
#14
Just tried changing Windows setting to comma for the decimal separator, and clearly reproduced the problem you reported :) I'll add a fix for this, so that you don't need to change the files.
Thanks for the future fix :) I have tried to change the files switching the commas and so on but wihtout luck (didn't report here cause I got busy), so it was a problem with your code (I even loaded them on REW and saved them again with no luck on your software indeed).

And have you had any issues with glitchy test tone (or any other type) while changing volume with the slider and the frequencies on the go?
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #15
What device are you using and at what
Thanks for the future fix :) I have tried to change the files switching the commas and so on but wihtout luck (didn't report here cause I got busy), so it was a problem with your code (I even loaded them on REW and saved them again with no luck on your software indeed).

And have you had any issues with glitchy test tone (or any other type) while changing volume with the slider and the frequencies on the go?
Can you please try downloading the updated file? I made a change that should help with loading files when decimal separator is mismatched. At least it worked fine on my test computer :) It's not a finished release of the next version, but I wanted to make sure this fixed the problem you reported. Also, please check if the sound continues to play when you exit the app, I think that problem should also be fixed.

https://distortaudio.org/EarfulSetup.zip
 

zermak

Senior Member
Joined
Jun 2, 2019
Messages
315
Likes
203
Location
Italy
#18
Downloaded and tested. The curves open properly now but while I am messing with the volume slider and the frequencies, the audio keeps glitching and the CPU spikes at 100%.
 
OP
pkane

pkane

Major Contributor
Forum Donor
Joined
Aug 18, 2017
Messages
1,451
Likes
2,153
Location
North-East
Thread Starter #19
Downloaded and tested. The curves open properly now but while I am messing with the volume slider and the frequencies, the audio keeps glitching and the CPU spikes at 100%.
The CPU spike may not be a big deal, but the sound glitch sounds like an issue. What speed/type of CPU do you have?

Edit: I actually don't see any CPU utilization change when adjusting volume on my test computer. What type of audio driver are you using (DS, WASAPI or ASIO)? This may also make a difference. Any DSP in the audio chain?
 
Last edited:

zermak

Senior Member
Joined
Jun 2, 2019
Messages
315
Likes
203
Location
Italy
#20
I have a i5 6600K and all the four physical cores are used when it freezes and has glitches.

I have to try again with other outputs, it happens with the standard Windows DS for sure and I am using EqualizerAPO for both headphones/IEMs and desktop speakers. But if I recall I changed output too and I think I had the same issue in WASAPI (exclusive and shared) but I didn't test properly and maybe it is just DS.
 
Top Bottom