I’ve tested the script again and it seems to be running fine for me. If you have this and other scripts running I would suggest you turn them off one by one to narrow down which one is the culprit.
For example I see you are running the memrise turbo script so it could be that.
Thanks so much for this!! Working for me today on Chrome 70 for Mac with Korean.
Do you have some kind of instruction manual to use your script to bulk upload forvo audio into my Hebrew-english course ? Thank you
https://www.memrise.com/course/2113613/eteacher-rosen-modern-hebrew-1-2-3/
Hi Renan,
Cooljingle’s script just adds TTS audio for the given language.
It does not bulk upload any audio or search for audio resources for the selected word.
Do you know some script that does ? Thanks
Hi @cooljingle
I just found out that the audio provider script V0.1.26 crashes for this course and its level 2 pronunciation either in the preview or learning mode.
Works great for the 1st or 3rd level with a normal Japanese definition.
TypeError: _.find(...) is undefined[Weitere Informationen] eval line 2 > eval:98:39
MEMRISE.garden.session.make_box</< moz-extension://0ae47eb2-9a98-49a8-bf05-561a74b06db8/userscript.html:98:39
MEMRISE.garden.session.make_box</< moz-extension://0ae47eb2-9a98-49a8-bf05-561a74b06db8/userscript.html:143:30
[22]</l<.activate_box https://static.memrise.com/learning/dist/js/learning-session-31ea512a7413.js:1:57584
[22]</l<.begin https://static.memrise.com/learning/dist/js/learning-session-31ea512a7413.js:1:57349
b https://static.memrise.com/learning/dist/js/learning-session-31ea512a7413.js:1:126379
x/<.success https://static.memrise.com/learning/dist/js/learning-session-31ea512a7413.js:1:126766
S.Callbacks/c https://static.memrise.com/dist/js/libs-a4fd8e8bf113.js:1:27470
S.Callbacks/h.fireWith https://static.memrise.com/dist/js/libs-a4fd8e8bf113.js:1:28219
u https://static.memrise.com/dist/js/libs-a4fd8e8bf113.js:1:77237
.send/o/< https://static.memrise.com/dist/js/libs-a4fd8e8bf113.js:1:79492
t https://static.memrise.com/dist/js/libs-a4fd8e8bf113.js:1:436024
onStopRequest resource://gre/modules/WebRequest.jsm:347:7 From Audio-Provider user script source, from the nested stack (1st error stacktrace line)
if (wordColumn !== "No audio") {
var isInjected = false;
if (!(canSpeechSynthesize || canGoogleTts || canVoiceRss)){
log("could not find a way to generate audio for language " + language);
$('#audio-provider-link').hide();
} else
isInjected = overrideAllAudio || (result.presentationData || result.testData).audio.value.normal === "AUDIO_PROVIDER";
currentWord = _.find([result.learnable.definition, result.learnable.item], x => x.label === wordColumn).value;Prevous “Auto learn” caller, the user script from the nested stack (2nd error stacktrace line)
MEMRISE.garden.session.make_box = (function() {
var cached_function = MEMRISE.garden.session.make_box
return function() {
var result = cached_function.apply(this, arguments);I think the problem is with the missing Japanese definition column on the pronunciation levels.
If this might be the case (you can tell better), why don’t you include an audio provider setting to omit the audio for any “Japanese pronunciation” column definition as given?
It should be somehow possible to plant all the pronunciation levels on the web portal and have your user script activated for the rest of the normal levels.
Is there any workaround way to modify the match header to exclude “wildcard (pronunciation) wildcard” level URLs?
I have suggested your user script here: [Feedback] Advanced Japanese Vocabulary course no longer displays phonetic kana readings in reviews - #7 by Thomas.Heiss
Hello, I’ve been using this script for several years and I wanted to know if you could adapt it (if possible) to the Memrise application. As a computer novice, I have no idea how to do it or if it already works!
Have you tried it with the Tampermonkey / Violentmonkey addons on the mobile web browser apps of Firefox or Chrome on your mobile device (smartphone or tablet)?
Camilo provided a list of compatible browser addons (I have never tested them on an Android emulator): https://github.com/camiloaa/duolingotreeenhancer
I installed the Tampermonkey application on my phone and the script is present. However, when I use the search bar integrated into the application and go to the memrise site, it’s impossible to access the learning or revision, in other words the addresses for which the script is supposed to work. Indeed, even if the tampermonkey logo warns me that the script is being used, the page can’t load despite a 15-minute wait… and if I use google or the application memrise, the script doesn’t work! Will this come from my phone?
Unfortunately there’s no way to make the script work with the memrise app. I haven’t tested using the script in a mobile browser but in theory it should work, although it may require you to set the browser to request the desktop version of the website.
Hi Thomas, I tried previewing level 2 of the course and the audio was working fine, let me know if there’s something else I need to do to reproduce the issue.
Hi @cooljingle, I’m very glad to have found your script collection. Amazing contribution, you should be hired by Memrise 
In the “Decks” feedback forum I have suggested @MemriseSupport to add audio automatically to the entries of the course. For me, it is painful to download all audios and upload to Memrise platform.
When you are learning a language, I think it is important to select carefully which dialect/variety you want to hear repeatedly in order to memorize and integrate into your knowledge. I couldn’t resist asking in the mentioned thread, but now I’m arising this question in your thread.
Is possible with your script to select which variety of English you want to hear? British RP, British Cockney, American, Scottish, South-African, etc.
Thank you very much for your time! 
Hey, unfortunately it doesn’t seem to work with memrise decks 
(with Chrome). Does anyone one what’s wrong?
Thank you so much for this script! It’s been extremely helpful for me. I’ve run in to a problem with Korean Hanja (Chinese characters that are the root of Korean words). The pronunciation it’s providing is incorrect because the same character can have different pronunciation is Chinese, Japanese, and Korean. For example 物 in Korean is pronounced 물 (mool), but this script produces a completely different pronunciation. Is there a way for the script to detect the intended language and use the pronunciation of of that language for the character?
Hey !
(sorry for my english)
firstly, thanks a lot for memrise audio provider, you made my dream come true !
currently i learn on my own course that comprises more than 4000 words, it was just impossible to record them one by one… 
i have a question though:
i changed the script “English”: “en-gb”, to “English”: “en-us” , (cause i need the us accent)
it works, i hear a female voice with a us accent, but it sounds not clear…
it’s kind of uncomfortable. a bit like a rusty robot 
is it possible to change it ?
ps:
in memrise settings, when i toggle my course to “us english for french” instead of “gb english for french”, the audio doesn’t work anymore
thanks for all !
kinds regard
Nils
I again have the same problem as the first poster here. Reviewing words comes to a stop and the whole thing freezes. I can save my progress but I need to start another review session. Only seems to be a problem when audio provider is on.
Sorry for the late reply. The script detects the language to use based on the course settings - so if it’s a course for korean it should be trying to speak korean, although it could be trying to pronounce the wrong ‘column’ e.g. the definition instead of the word itself, this you can change in the script settings:
if you still need help let me know the course URL and I can take a closer look.
Seems to still work for me - are you saying it used to work and at some point it stopped working?
Hi, thanks for the reply, even if late. I still use this plugin every day. The words being poronounced were all in the left word column, not the definition.
However, last week I noticed that the pronunciations are suddenly all correct! I looked around to see if anything had been updated, but couldn’t find any notes. Anyway, I belive this is fixed now.
