Tom Bellman

I am creating a basic pure tone audiometry test on max for my dissertation and I have the basic skeleton, just not quite there yet.

I have a couple of questions. Firstly, I want to be able to see exactly what dB level my objects, and therefore my laptop, are outputting so I can record this for my data collection. I have an 'atodb' object but that only records the dB output of the objects but I can obviously turn my laptop up and down and these numbers would not change, any suggestions? I also need the level's to go up in 5dB steps, is this possible without guessing where to click?

Secondly, I am using a USB button to record my participants responses, so they click if they can hear it and don't click if they can't. I want to use a 'coll' object which basically writes 1's (when they click) into an excel sheet in specific places but I have no idea where to start with this.

Would really appreciate some help with these things!

Please see the patch I have so far attached.

I am creating a basic pure tone audiometry test on max for my dissertation and I have the basic skeleton, just not quite there yet.
I have a couple of questions. Firstly, I want to be able to see exactly what dB level my objects, and therefore my laptop, are outputting so I can record this for my data collection. I have an 'atodb' object but that only records the dB output of the objects but I can obviously turn my laptop up and down and these numbers would not change, any suggestions? I also need the level's to go up in 5dB steps, is this possible without guessing where to click?
Secondly, I am using a USB button to record my participants responses, so they click if they can hear it and don't click if they can't. I want to use a 'coll' object which basically writes 1's (when they click) into an excel sheet in specific places but I have no idea where to start with this.
Would really appreciate some help with these things!

Please see the patch I have so far attached.


pure-ton-audiometry-test-on-max

Hmm if I understand you correctly you're looking for something that's not really possible.

Max cannot know how many dB of sound output your system actually provides. That depends on the following signal chain.

Notice that the dB Max shows you internally are dB fullscale, measured only in the in the digital domain. Although dB Fullscale and dB SPL (sound pressure level) sound similar they are two very different things. You would need to calibrate the actual output of a system with a special microphone to really now the output in dB SPL

Hmm if I understand you correctly you're looking for something that's not really possible.
Max cannot know how many dB of sound output your system actually provides. That depends on the following signal chain.
Notice that the dB Max shows you internally are dB fullscale, measured only in the in the digital domain. Although dB Fullscale and dB SPL (sound pressure level) sound similar they are two very different things. You would need to calibrate the actual output of a system with a special microphone to really now the output in dB SPL


I thought this would be the case, I will have to go back to the drawing board with this one. Maybe somehow measure the SPL that my headphones are outputting so I can scale it on Max.

Have you any ideas about the other points?

I thought this would be the case, I will have to go back to the drawing board with this one. Maybe somehow measure the SPL that my headphones are outputting so I can scale it on Max.
Thank you for your help!
Have you any ideas about the other points?


as first switching cycles~ like that will not be helpfull.

You have to turn them on or off, or just use one cycle with different

frequencies and pack them into Freq/level lists for storage.

Making 5 db steps is easy to do, but is that a good way ?

It is big difference if we go from -70 to -75 db, or from

The scale should be finer, if You need to do this precisely.

You can even make a freq sweep, which user could stop

Storage into coll is also no problem, one can use

frequency as Address and Level as Value when Freq gets heard.

Hi Tom,
as first switching cycles~ like that will not be helpfull.
You have to turn them on or off, or just use one cycle with different
frequencies and pack them into Freq/level lists for storage.
Making 5 db steps is easy to do, but is that a good way ?
It is big difference if we go from -70 to -75 db, or from
-5 to 0 db in Max.
The scale should be finer, if You need to do this precisely.
You can even make a freq sweep, which user could stop
using the switch.
Storage into coll is also no problem, one can use
frequency as Address and Level as Value when Freq gets heard.


I am only going off what I have read previously about the pure tone audiometry tests and they work in 5-10dB steps.

Would you say there is an easier way? Sorry to be a pain but could you demonstrate this on my patch and send it back to me?

I am only going off what I have read previously about the pure tone audiometry tests and they work in 5-10dB steps.
Would you say there is an easier way? Sorry to be a pain but could you demonstrate this on my patch and send it back to me?

Sure, here is a simple patch with ramp from -120 - 0. db in 10000 ms.

When user triggers the button, ramp stops, and

If You want I can post 5db step up version ...

And maybe add left / right ear meassurement separately ?

Sure, here is a simple patch with ramp from -120 - 0. db in 10000 ms.
You can change that as You wish.
When user triggers the button, ramp stops, and
freq and level gets stored in coll
-----------
If You want I can post 5db step up version ...
And maybe add left / right ear meassurement separately ?


if the PA for the setup is already available to you, you could just measure it and memorize the data in your app, for some kind of gain to phon conversion, display or control.

this is how you can also vaguely "calibrate" a contourfilter function for a big concert PA and similar things.

about your question with the 5db steps - there is dbtoa, the opposite of atodb which wil do this for you. :)

This patch is perfect, although I am going to try and re-create it in a slightly different way as I do not want to copy your work and need to make something myself, just needed some guidance.

Just need to get my head around how this works now.

Thank you so much Source Audio!
This patch is perfect, although I am going to try and re-create it in a slightly different way as I do not want to copy your work and need to make something myself, just needed some guidance.

Just need to get my head around how this works now.
Again, thank you so much for your help.
Tom


Hi, Tom, we are all here on the forum for getting anf giving help or advice, and that's all about it.

So please, just ask if there is anything else that could be worked out together.

Hi, Tom, we are all here on the forum for getting anf giving help or advice, and that's all about it.
There is no my work your work.
So please,  just ask if there is anything else that could be worked out together.


Great, thank you so much again for your help!

I want my patch to match the official pure tone audiometry test as closely as possible.

So i want to be able to select certain dB levels before playing the tone, and then select the next level and frequency etc.

I imagine this is simple but I can't seem to work it out. Some pointers as to how this is done will be much appreciated.

Great, thank you so much again for your help!
I want my patch to match the official pure tone audiometry test as closely as possible.
So i want to be able to select certain dB levels before playing the tone, and then select the next level and frequency etc.
I imagine this is simple but I can't seem to work it out. Some pointers as to how this is done will be much appreciated.

Hi Tom, that should be easy to implement.

You could use menu with set levels in db,

One could step through levels up or down using

Selected Level will than get stored in coll, instead of interrupted

If You need I can make a patch for that, but please tell me what

should level scale look like, 5 db steps ?

You coulkd of course just use one Float with arbitrary db scale,

Hi Tom, that should be easy to implement.
You could use menu with set levels in db,
and Play / Stop button if You wish.
One could step through levels up or down using
for example up/down arrow keys.
Selected Level will than get stored in coll, instead of interrupted
ramp.
If You need I can make a patch for that, but please tell me what
should level scale look like, 5 db steps ?
Starting from ?
You coulkd of course just use one Float with arbitrary db scale,
but that would not be really practical.


I have tried to put something together but not sure I am doing quite the right thing. Would be excellent if you could give me an example?

Lets say from 5dB up to 70dB in 5dB steps if possible.

Really appreciate your help, and I am going to try and re-build what you have made just so that I totally understand it.

I have tried to put something together but not sure I am doing quite the right thing. Would be excellent if you could give me an example?
Lets say from 5dB up to 70dB in 5dB steps if possible.
Really appreciate your help, and I am going to try and re-build what you have made just so that I totally understand it.

but the way it works in Max is different than producing

There is no way to output 5 db of sound pressure from Max , because we don't know

So for range of 65 db, we just start from-65 db and clime up in 5 db steps up to 0 db,

Then it is another task to calibrate/adjust the external gear.

Ok, I'll post something soon,
but the way it works in Max is different than producing
sound pressure in decibels.
There is no way to output 5 db of sound pressure from Max , because we don't know
what Yor amp/speakers/phones etc do.
So for range of 65 db, we just start from-65 db and clime up in 5 db steps up to 0 db,
which is maximum level.
Then it is another task to calibrate/adjust the external gear.


No no of course I understand that. The output is in the digital domain. That's my next challenge!

I think i am going to be doing my test on a pair of headphones so will just have to either use an SPL level really close to the headphones to determine a level which is consistent for all of my tests, or use speakers in a studio and keep a consistent level.

You have been an absolute saviour! Thank you for all of your help, I do appreciate it.

Wish I could help you out in some way in return!

No no of course I understand that. The output is in the digital domain. That's my next challenge!
I think i am going to be doing my test on a pair of headphones so will just have to either use an SPL level really close to the headphones to determine a level which is consistent for all of my tests, or use speakers in a studio and keep a consistent level.
You have been an absolute saviour! Thank you for all of your help, I do appreciate it.
Wish I could help you out in some way in return!

And I also received a lot of help and tips here ...

You are welcome !
And I also received a lot of help and tips here ...


How did you get the arrows to control the Level Umenu? I just want to test having the left and right arrows controlling the frequency select.

One final question.... Sorry!
How did you get the arrows to control the Level Umenu? I just want to test having the left and right arrows controlling the frequency select.

If You doubleclick the "keys-step" patcher You will see it all.

Key Codes for left and right arrows are 28 and 29.

But maybe it is easier with updated patch:

If You doubleclick the "keys-step" patcher You will see it all.
Key Codes for left and right arrows are 28 and 29.
But maybe it is easier with updated patch:


Pure ton Audiometry test on Max