HTML folder not generated, "decoding audio data failed"

[s.bassoon]

URL of experiment: Sign in · GitLab

Description of the problem: The experiment runs fine locally. When I sync or Export HTML, an index.html file is generated, but an HTML folder is not generated:

I have tried moving the experiment folder to a different location on the same drive, and to a different drive altogether, but still no HTML folder is generated.

When I sync to Pavlovia, all appears well:

Pulling changes from remote…
Already up to date.
done
Pushing changes from remote…
done
Successful sync at: 10:00:17, took 1.762s197.5659 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\verbalLoadCondsPractice.xlsx as conditions, 4 conditions, 12 params
197.5799 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\chooseBlock.xlsx as conditions, 2 conditions, 1 params
197.5855 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\sepConditions.xlsx as conditions, 4 conditions, 6 params
197.5923 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\combConditions.xlsx as conditions, 4 conditions, 6 params
197.6033 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\chooseBlock.xlsx as conditions, 2 conditions, 1 params
197.6087 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\sepConditions.xlsx as conditions, 4 conditions, 6 params
197.6152 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\combConditions.xlsx as conditions, 4 conditions, 6 params
197.6373 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\verbalLoadConds.xlsx as conditions, 120 conditions, 12 params
198.6847 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\verbalLoadCondsPractice.xlsx as conditions, 4 conditions, 12 params
198.6984 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\chooseBlock.xlsx as conditions, 2 conditions, 1 params
198.7039 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\sepConditions.xlsx as conditions, 4 conditions, 6 params
198.7103 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\combConditions.xlsx as conditions, 4 conditions, 6 params
198.7212 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\chooseBlock.xlsx as conditions, 2 conditions, 1 params
198.7267 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\sepConditions.xlsx as conditions, 4 conditions, 6 params
198.7330 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\combConditions.xlsx as conditions, 4 conditions, 6 params
198.8320 EXP Imported D:\OneDrive - MMU\research\Online Stroop\Verbal Load\verbalLoadConds.xlsx as conditions, 120 conditions, 12 params
202.4519 INFO Successful sync at: 10:00:17, took 1.762s

However when I attempt to run the experiment I get a series of errors like this. I assume this is because there is no HTML folder?

Strangely, if I then subsequently ‘sync with web project’, I am prompted to create a project because a project does not exist, despite the project still appearing on the Pavlovia dashboard

[JensBoelte]

Hello,

which version are you using? In more recent versions, no html-folder is being generated. You find all resources in a folder called resources. The link to your project does not work (error 404).

Best wishes Jens

[s.bassoon]

Hi, thanks for the reply.

I created the experiment in 2020.2.5 but then updated to the latest 2021.1.4 to see if that resolved the issue.

Ok thanks – although this folder is also missing (see first screen shot)?

That’s odd re the link, when I click on it, it takes me to the experiment

[JensBoelte]

Hello,

the screen-shot is from your PC? The folder resources is created on git. Your project is probably not public and therefore only accessible for you.

Best wishes Jens

[s.bassoon]

ok thanks, sorry I didn’t know i needed to change the visibility, I think this should be public now.

Looking at the gitlab folder, there is no ‘resources’ folder here:

[JensBoelte]

Hello,

change the backslash in your excel-files to a forward-slash. BTW, it is rather difficult to determine which files your program is using given that you have copies of files in different directories (you even have zip-file in your directory). I’d get rid the irrelevant directories to understand better what is being presented. Your program directory should contain only what your need to run the experiment.

In addition, some files are sampled with 48.000 Hz, others are sampled with 44.100 Hz. You might want to use the same sampling rate for all files. See here for some suggestions about the sample rate.

Best wishes Jens

[JensBoelte]

Hello,

you are loading all stimuli via excel. Right? Then don’t care about a resources folder.

Best wishes Jens

[s.bassoon]

Yes that’s correct

[s.bassoon]

Thanks a lot for this

change the “” in your excel-files to “/”.

I guess you mean change “backslash” to “/” right? Edit, look like this text editor removes the backslash symbol, so I’m sure that’s what you meant.

I’ve replaced “backslash” with “/” in the verbal load spreadsheets, deleted the previous project from Pavlovia and started again with a tided directory.

1. I ‘sync with web project’
2. Create a new project on Pavlovia

Sign in

3. Run pilot

Same error

[wakecarter]

Some browsers don’t like wave files which is why mp3 is recommended for online use

[s.bassoon]

I’m currently running this in Chrome on Windows. I have another audio experiment on Pavlovia using .wav files which runs OK, do you think it is worth converting everything to mp3 to check?

[wakecarter]

Is it always the same file? If other files work then you might just need to recreate that file.

[s.bassoon]

Hi, no it is every audio file in the experiment.

I think I have resolved this issue:

I specified 64 bits per sample when I created the stimuli, a legacy from a previous experiment that required this number of bits. So I have recreated all of the stimuli with 16 bits per sample, to see if this makes a difference.

However, when I try to sync the project I get this message again:

Looking at some previous threads from last year, this seems to have been a wide spread issue, but I’m not sure it is was ever resolved?

So I had to delete the old project from gitlab and create a new project…

With the reduced bits per sample the experiment now runs ok.

Thanks for everyone’s help. Was there ever a solution for the issue with having to recreate a new project each time I sync with gitlab? (should I create a new thread?)

[wakecarter]

Where are you saving your local experiment? I think that this can be caused by saving to a cloud drive (which might mean that the hidden .git folder isn’t automatically available to tell PsychoPy the location of the project online).

[s.bassoon]

Ah ok, it’s saved to a OneDrive folder (so I can sync with my work PC). If I copy to the C:/ drive it syncs with gitlab fine.

Ok thanks. That’s slightly irritating, but at least I know what the problem is!

Thanks again for your help

[wakecarter]

Remember that there’s a copy on gitlab so you could also sync from your work computer.

However do reopen the Builder file after a sync where your last change was from another computer