Google text-to-speech error

[W NW]

Dear,

I am using 2.7.3 Mnemosyne version under Windows 7.

Since couple of days I have the below error message when trying to add the google text-to-speech: "An unexpected error has occured. Please forward the following to the developers: 

Traceback (innemorst last): File "mnemosyne\pyqt_ui\qtextedit2.py", line 107, in keyPressEvent (...)" etc. - I paste the screen.

Effectively no Google-pronounciation is added to cards; before it used to work perfectly under 2.7.3.

Could you please let me know any suggestions?

regards

blickwn

[Peter Bienstman]

Hi,

I'm afraid it works fine on my Windows 10 machine. It might be a temporary glitch with the Google servers. So far I have not received any other reports from this. Note that Windows 7 is no longer supported, it reached end-of-life in January 2020...

Cheers,

Peter

--
You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/b21c00fb-7f2c-42a8-be51-215f740e8b96n%40googlegroups.com.

[W NW]

Hi, thanks for taking time. 

Under Windows 7 I have been using Mnemosyne 2.7.3  since it was published with no issues at all. So it would be quite strange that Win7 is the main culprit , though of course one cannot exclude it. 

Today I noticed two similar errors, please see below. I also noticed that the issue occurs also when trying to export cards - after clicking the last "OK" before exporting; and the cards are not exported. 

Today I tried uninstalling and installing the Mnemosyne again - with no effect. Would you have any other suggestions please? The text-to-speech is a very important function to me, used on a daily basis...

regards

blickwn

[Oisín Mac Fhearaí]

The "lang" parameter in the call to gTTS seems to have caused problems for people recently.

See this example: https://github.com/pndurette/gTTS/issues/252

The problem in the cases I've seen were caused by a language string of "en-US" and "en-CA". In both cases, switching to just "en" fixed the problem.

Looking at the file google_pronouncer.py in the Mnemosyne source suggests that the language parameter is derived from the card type property "sublanguage_id", or the property ""language_id" if that's not available. Can you check if they're set to something like "en-US" and replace them with "en" to see if that helps?

Oisín

To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/460ee7bc-bc46-4466-92b2-d7fff4b08037n%40googlegroups.com.

[Peter Bienstman]

Thanks for tracking this down! (My own cards never used a sublang, which is why I could not reproduce it.)

I noticed that the people behind gtts, the google text-to-speech library, are preparing a new release to fix these and other issues, and when it's released, I can prepare a new Windows installer.

Meanwhile, Oisin's workaround of not selecting a sublanguage should help.

Cheers,

Peter

To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/CAO-0pXC%3DUvaUfb4iVWhr7V1ZRQ9%3D%2Bb_12Sh_nA2fTAShyqhcTQ%40mail.gmail.com.

[W NW]

Thank you Oisin and Peter. 

I am afraid that I will have to hope for the Windows installer that Peter mentioned as I am not "technical" enough to change the Mnemosyne source code... I also do not know how not-to select the sublanguage.

Or perhaps I am missing something? And if the workaound could be applied from the level of Mnemosyne itself or by editing some Mnamosyne files -  could you please give me more details how to do this?

regards

blickwn

[Peter Bienstman]

In the text-to-speech dialog there's an option to choose the sublanguage in case more than one variant is available. Just make sure it's set to 'default'.

Cheers,

Peter

To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/008d49af-3104-440d-8bd8-9f0c2d13ea2an%40googlegroups.com.

[W NW]

OK,  thanks - I see four variants for English language: UK, US, India, Australian, please find screen below; but whichever I choose I have the same error, and it appears the moment I am choosing one of the variants (ie. even without clicking Preview or Insert). What's more - the text-to-speech dialogue appears only when the Back of the card   (= my foreign language) is empty; and when there is a word written in the Back of the card I even cannot get to the text-to-speech dialogue because the error occurs immediately when I choose Ctrl+P or Insert GTTS from the menu. 

Am I doing something wrong here?

Just for the full picture I attach also the screen of my Card-types configuration; 

the same error concerns both of my card types: English and French (with French having  two language variants: France and Canada).

regards

blickwn 

 

[Peter Bienstman]

OK, let's see if gtts quickly releases a new version, otherwise I'll hack together a temporary workaround.

Peter 

To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/0e6087ba-c221-4f47-9daa-e7a461c054e7n%40googlegroups.com.

[Peter Bienstman]

Try this version: https://www.dropbox.com/s/fjge0mdj52cckao/mnemosyne-2.7.3b-setup.exe?dl=0

Peter

[W NW]

Dear Peter, 

this version works perfectly, gtts and also exporting. I am really impressed by the support here, thank you so much!

regards

blickwn