Hello everybody,
I am working with audio-based Implicit Association Tests (IATs). Because we have a huge amount of different audio files, we passed the upload limit for the web script when using audio files in the .wav version. That's why we changed our script and exclusively used .mp3 audio files to avoid to pass the upload limit.
However, we encountered a problem when using .mp3 files instead of .wav files. We are aware of the fact that we should use the <video> command when working with .mp3 files, but we are working with different randomization trials which seems less feasible when using the video command instead of the sound command.

Interestingly, the script (Inquisit 5) worked without any errors after having replaced all .wav file endings by the file ending .mp3 on our MAC computers (only replacing the endings, no further changes were made), but NOT in WINDOWS. The WINDOWS systems showed different errors (e.g xyz is not a .wav file, or cannot locate xyz).
Because our script works perfectly with the .wav endings, we wondered whether it might be possible to adjust the functions in MAC and windows to make it possible to play .mp3 file endings in the WINDOWS versions. It is important for our study that the script works both in MAC and Windows versions.
I am looking forward to hearing from you
Thank you for your time

For audio formats other than WAV, you must use the <video> element for things to work on both platforms. I'm having trouble understanding the following:

"We are aware of the fact that we should use the <video> command when working with .mp3 files, but we are working with different randomization trials which seems less feasible when using the video command instead of the sound command."

Could you please explain why you think that is / how you came to that conclusion? <sound> and <video> elements are stimulus elements just like <text> and <picture>. Randomization-wise, they all essentially work exactly the same. I.e., it is not obvious at all why using <video> as opposed to <sound> would make a difference.

Thank you for your reply.
We are working with different speech samples from different speakers, so each participant will be presented different speech samples. Therefore, we used a randomization in a way that e.g. the items iatlufr will be different for each participant (e.g. for participant 1 => iatlufr setiatfrlu1). This approach worked perfectly when using WAV and sound commands, however when we replaced the WAV endings by MP3 endings, the script showed the errors described above.

For instance, the following command shows red error warnings, but it worked with the sound command.
<video attributejos>
/items = iatlufr
/timeout = 10000
</video>

We are surprised that the script (with the replacement of WAV endings by MP3 endings) works perfectly when using MAC and it shows errors and does not play the script when using WINDOWS.
My apologies for my description that may cause misunderstanding, but I am not an expert in the field of INQUISIT, but I really enjoy conducting experiments with INQUISIT and would be really grateful if you could give me sont advice on my issue.
Or is it even possible to check whether there are differences between MAC and WINDOWS with respect to the use of MP3 files?
Thank you for your time

An example of our randomization:
<variables>
 /group = (1 of 100 (setlernlulu = setlernlulu1; setevalulu = setevalulu1; setlernlufr = setlernlufr2; setevalufr = setevalufr2;setlernfrlu = setlernfrlu1; setevafrlu = setevafrlu1;setlernfrfr = setlernfrfr2; setevafrfr = setevafrfr2; setapplulu = setapplulu1; setapplufr = setapplufr2; setappfrlu = setappfrlu1; setappfrfr = setappfrfr2; iatlulu = setiatlulu1;iatlufr = setiatlufr2;iatfrlu = setiatfrlu1;iatfrfr = setiatfrfr2)
 /group = (2 of 100 (setlernlulu = setlernlulu2; setevalulu = setevalulu2; setlernlufr = setlernlufr1; setevalufr = setevalufr1;setlernfrlu = setlernfrlu1; setevafrlu = setevafrlu1;setlernfrfr = setlernfrfr2; setevafrfr = setevafrfr2; setapplulu = setapplulu2; setapplufr = setapplufr1; setappfrlu = setappfrlu1; setappfrfr = setappfrfr2; iatlulu = setiatlulu2;iatlufr = setiatlufr1;iatfrlu = setiatfrlu1;iatfrfr = setiatfrfr2)
......
 

The "red error warning" you get has most certainly nothing to do with switching <sound> to <video>.
#1: Neither <video> nor <sound> elements have a /timeout attribute. This will not lead to a "red error warning", but a simple "yellow warning" saying
"/timeout is not a known attribute."
#2: The syntax in your <variables> element is faulty. That is the likely reason for any errors you receive.

<variables>
 /group = (1 of 100 (setlernlulu = setlernlulu1; setevalulu = setevalulu1; ...)
 /group = (2 of 100 (setlernlulu = setlernlulu2; setevalulu = setevalulu2; ...)
...
</variables>
2a: You are missing parentheses.
2b: The proper separator is a comma, not a semicolon.

I.e. the above needs to read

<variables>
 /group = (1 of 100) (setlernlulu = setlernlulu1, setevalulu = setevalulu1, ...)
 /group = (2 of 100) (setlernlulu = setlernlulu2, setevalulu = setevalulu2, ...)
...
</variables>

#3: Expressed simply, the reason why you ought use <video> is the following: The OSX media-subsystem is such that it (tries to) figure out the type of file being played internally. I.e., on a Mac, <sound> and <video> ultimately map to the same, single subsystem. Not so under WIndows: Windows treats raw WAVs and other media formats differently. They are handled by two separate subsystems, if you will, and the one dedicated to WAVs cannot handle other media formats.

Hope this helps.

Thank you for your reply.
I changed the syntax in <variables>.

However, there are still errors (red) popping up for the following paragraph (which works in a script using WAV files and sound commands):

<video soevafrfr>
/items = setevafrfr
</video>

<video setevalulu1>
/items = ("A-Yannick-refresh.mp3")
</video>

=> error: Could not locate item 'setevafrfr'.

Do you have an additional advice?
Thank you for your time

The error tells you that Inquisit is unable to locate the <item>  element referenced in

<video soevafrfr>
/items = setevafrfr
</video

There is still some mistake in the script. Most likely, it is in your <variables> element. Double-check its various /group attributes and make sure things are assigned correctly and that there aren't any typos.

Thank you. I will have a look at it.

If you can't find the mistake, you can attach the script [1] to this thread (click '+Insert -> Add File') when posting a reply. I'll happily take a look / help with locating the error.

[1] The *.iqx file is enough, there is no need to attach any of the stimuli (e.g. mp3 files).

Hi again,
thank you for your reply and the your offer to have a look at the script.
I tried to correct the errors but the script still won't work because there are still some error warnings like "could not locate the element" or "unable to open file..."
I think the problem occurs due to the randomization we created in <variables>
If you could have a look, that would be very helpful! Maybe you've got some advice which commands to use in our study.
Thank you very much in advance

Your syntax for the various <item> elements is completely botched. Please allow me to point out that this has *nothing* to do with using <video> instead of <sound>. You would encounter the *exact* same problem if you were to use that botched syntax with <sound> elements.

<video solernlulu>
/items = ("setlernlulu")
</video>

is not how you reference an <item> element. This needs to read

<video solernlulu>
/items = setlernlulu
</video>

Your <variables> element determines *what that reference means*. I.e., you do not actually create an <item> element called <item setlernlulu> in the script. Instead, your <variables> determine which other <item> element to *treat* as "setlernlulu".

In other words: Apart from the also completely botched syntax, the below does not make sense at all.

<item setlernlulu>
/1 = ("setlernlulu1")
/2 = ("setlernlulu2")
/3 = ("setlernlulu3")
/4 = ("setlernlulu4")
</item>

What you do is this:

#1: Specify

<video solernlulu>
/items = setlernlulu
</video>

#2: Create four item elements

<item setlernlulu1>
/ 1 = "A-Yannick-Lernaufgabe.mp3"
</item>

<item setlernlulu2>
/1 = "A-Denis-Lernaufgabe.mp3"
</video>

<item setlernlulu3>
/1 = "A-Yves-Lernaufgabe.mp3"
</item>

<item setlernlulu4>
/1 = "A-X-Lernaufgabe.mp3"
</item>

#3: In your <variables> element's /groups, specify which of the above <item> elements "setlernlulu" is supposed to refer to:

<variables>
 /group = (1 of 100) (setlernlulu = setlernlulu1, ...)
 /group = (2 of 100) (setlernlulu = setlernlulu2, ...)
...
</variables>

Do the exact same thing for the other <video> elements.

Dear Dave,
thank you for your reply,
I tried a lot of different variations of syntax leading to this botched syntax. My apologies for that but I did not know how to handle the randomization specified in the <variables> section.
I will have a look at your comments and try to implement them in my script.
Thank you very much for your time and your comments.
Best wishes

Hi there,
the script works now. My apologies for any inconvenience caused.
I know get that the script was completely batched and I will try to avoid such stupid mistakes in the future.
Thank you very much for your help and you explanations.
Have a good day
Best wishes