Memrise card template [support thread]

[Eltaurus]

Not in the current version, for the reasons I outlined here.

But since you are the second person asking for this feature, I’ll keep it in mind for future updates.

[Leonel]

Thank you very much for responding, have a nice day.

[butter_muffin]

Hi @Eltaurus, just found out about the update and I have been trying it out this week. I’m so happy about the tapping features for grammar XD Thank you so so much! They have really made my decks from Glossika upto another level.

The one thing I have been stuck and couldn’t trouble shoot on my own is the cloze deletion. I “sentence mine” using Language Reactor and thus I wanted to upgrade the template from the original one that Language Reactor offers to the cloze template.

This is what i get:

How do I populate the choices field without getting this disaster?

[Eltaurus]

@butter_muffin

How do I populate the choices field

I’ve updated the Add-on on AnkiWeb, so it should now process Fields with clozes properly. Try upgrading and refilling your Choices (using the “Overwrite” option).

@butter_muffin

I “sentence mine” using Language Reactor and thus I wanted to upgrade the template from the original one that Language Reactor offers

Judging by the screenshots, it looks like that extension places all the information into the same Field, resulting in a mess. But I’m not familiar with it, so it’s hard to say how it can be fixed. You’ll have to provide more info on the actual Field contents of the Cards and what steps you are following to switch the Note templates if you need any assistance with that.

[tienppham2]

Thank you for everything you have done for us, I followed all your instruction but Can you tell me where i can turn on countdown of front card. I tried to change

to
</30>
but it has no use, can you help me to do like this gif, countdown timer

[sorata]

you should surround code with backticks, we can’t see them otherwise.

[tienppham2]

ok it’s here
"

"
I change to
“on" “30s”

[Eltaurus]

The timer was never actually implemented in the template, because there was no interest in it, while other features were kept being requested. I’ll put it back into my todo list now.

It probably won’t be the same design as the one in your screen recording though, because I couldn’t come up with a way to generalize it on multiple-choice cards. The second style from the demo seems more practical to me at this point:

While it’s not as unintrusive, in my opinion, it still beats the original Memrise design, which required constantly flicking eyes to the very corner of the screen and back to keep track of the time while typing.

If you’re up for it, there is also a prototype code, distilled from my other card template, which you can use to set up the timer. If you tell, what kind of behavior you expect, I’ll be able to provide more detailed instructions.

[tienppham2]

Thank you! God bless you!

[ultraleetj]

Hi, probably dumb question.. but programming often times goes over my head…
I am a blind user of anki and have been creating materials with audio, images and text for a few days with it (I know, probably too impatient) and found this comprehensive template/add on. Accessibility on windows is OK, though it can always improve some but even though its not all the way there, its doable. For now, my use case is that I want to make some cards with music materials to review intervals. SO I liked the cloze application of this. I thought it was possible to generate several single choice questions and cloze cards from one given text using this? meaning each cloze (option I got in {{}}) generates a choice, is this right?
For reference, in the “text” field I had put:

This interval is: {{c1::second}} {{c2::minor}} {{c3::major}} {{c4::third}} {{c5::fourth}} {{c6::tritone}} {{c7::fifth}} {{c8::sixth}} {{c9::seventh}} {{c10::octave}}.

In the “extra” field which I guess is used for the correct answer, I typed in the answer, such as “major third”. However I am not sure if the correct answer is just one word, for example “octave” how does this behave if you answer correct (by selecting octave) and incorrectly (by selecting say, tritone and minor)? or the other way around, if the answer is two words not one would this still work?

So now what i want to do with this is to answer using all of the three input types and let this randomly select an input type every time each card is presented… or is that too advanced to implement and am I better of creating each card repetatedly with a single input type? I guess the batch editing add on can simplify this though. Or is this done using nestedlevels of cloze, like {{1::type{{2:tap{{3::write}}}}}} ? in which case I should type the same things three times and then configure it from the mode in the code? like this? {{1::minor{{2:minor{{3::minor}}}}}}
If I can automate some of this, much better because there are 144 cards I would have to make and if I need to specify the type for each one of them I would have to make 432 of them.
Hope this was clear.

Also the fill choices option… how does this work in practice? could anyone provide a really simple explanation? I understand you could create say, 10 cards lets say a question is a number and an answer is a letter, so from 0 to 9 and answer are from A to J. If I select all of these and use the fill choices option then… it would present all of the cards at random or would it convert the front and backs of the cards to multiple choices? assuming I made 144 cards then would it be listing 144! choices?

[Eltaurus]

@ultraleetj

it would present all of the cards at random or would it convert the front and backs of the cards to multiple choices?

“Fill Choices” is but a workaround for the fact that Anki does not allow one Card to directly access information from other Cards. This function simply gathers info from the answer Field on the neighboring Cards and puts it into the Choices Field on each Card, so that they could be used as meaningful incorrect choices. As such, this function does not change how the Cards are presented or convert the Cards themselves in any way. It only allows multiple-choice Cards (if they are enabled by the setup, such as the ...MultipleChoice+Typing preset from the shared deck) to be generated in addition to other Cards.

@ultraleetj

assuming I made 144 cards then would it be listing 144! choices?

No, the “Fill choices” only forms the pool of potential incorrect answers from which a Card will randomly draw 5 on each review (this number can also be adjusted, if required).

@ultraleetj

you could create say, 10 cards lets say a question is a number and an answer is a letter, so from 0 to 9 and answer are from A to J

Getting back to this example, when you use “Fill Choices” on these Cards, it will generate 9 potential choices for each Card (excluding the correct answers for each Card), so the potential choices for Card 1 will be B, C, …, J; for Card 2 they will be A, C, D, … J; for Card 3: A, B, D, …, J; and so on. When Card 1 is reviewed, out of those 9, random 5 will be selected and shuffled with the correct answer. This way, the Card may present the Choices H, D, A, J, C, B, for example, and the next time the choices might be I, G, A, B, F, C, and so on.

@ultraleetj

if the answer is two words not one would this still work?

Yes, multiple-choice Cards can handle multiple-word choices. The only thing prohibited in the answers is the pipe character (|).

---

---

@ultraleetj

I thought it was possible to generate several single choice questions

I’m not sure, what you mean by “single-choice questions”.

@ultraleetj

each cloze (option I got in {{}}) generates a choice, is this right?

Each closed part generates a potential choice, yes.

@ultraleetj

In the “extra” field which I guess is used for the correct answer, I typed in the answer, such as “major third”.

That’s not correct. The “Extra” Field is there only to provide additional info – it is not involved in evaluating the answer in any way (unless you changed the setup).

For the Cloze Cards, correct answers are determined by the same “Text” Field that is used for questions (Adding/Editing - Anki Manual). For example, on the second Card, in your case, you will be prompted with

second [...] major third fourth tritone fifth sixth seventh octave

and the omitted minor will be the expected correct answer.

I’m not entirely sure what you are trying to achieve with this setup, though, and whether Cloze Cards are really suitable for your purpose. If you provide more details on your logic behind this, I might be able to give better explanations.

---

---

@ultraleetj

So now what i want to do with this is to answer using all of the three input types and let this randomly select an input type every time each card is presented…

This is doable, you will need to paste the following code in the “user prior scripts” section on the Front template:

inputModes = ["typing", "mchoice", "tapping"];
if (!document.getElementById('backwrap')) {
  inputMode = inputModes[Math.floor(Math.random() * inputModes.length)];
  sessionStorage.setItem("input-mode", inputMode);
} else {
  inputMode = sessionStorage.getItem("input-mode") || inputModes[0];
}
document.querySelector(".card-content.front").setAttribute("mode", inputMode);

I’m not sure what you are expecting to achieve with the “tapping” in your case, however. With your example above, this mode wouldn’t produce meaningful tests, because it is intended to be used with Cards, where correct answers are long sentences.

---

---

@ultraleetj

am I better of creating each card repetatedly with a single input type?

This will work too. In some way, it will be better, because this won’t get the review history of different input types mixed. As typing is typically harder than multiple-choice, for example, Cards with different input modes will tend to have different review intervals, which could naturally emerge in Anki if the Cards are kept separate. Randomly selecting a mode can lead to cases where you correctly answer a Card via multiple-choice, the interval gets increased, and then the Card gets too hard if you get it next time in the “typing” mode. If you fail it, and get it as a multiple choice next time it appears, it might be too easy again, and so on.

The implementation should be a bit different from what you described, however.
First, you don’t need to repeat the clozed part several times; only cloze the same answer thrice:

{{c1::{{c2::{{c3::minor}}}}}}

(instead of {{c1::minor{{c2::minor{{c3::minor}}}}}}, otherwise the correct answer for the first Card will be minorminorminor, and not what you expect).

Second, if you are still set on using tapping cards with your material, you will need to specify what cloze number corresponds to what input in the template, by entering the following into the “mode” attribute:

{{#c1}}typing{{/c1}}{{#c2}}tapping{{/c2}}{{#c3}}mchoice{{/c3}}{{#c4}}typing{{/c4}}{{#c5}}tapping{{/c5}}{{#c6}}mchoice{{/c6}}...

(and so on up to the maximum number of clozes you plan to use). This part is only a one-time setup, so it shouldn’t be that much trouble.

---

---

@ultraleetj

Accessibility on windows is OK, though it can always improve some

I’m always open to suggestions. If you have advice on how to make the template more accessible, it will be appreciated.

[ultraleetj]

Thanks for the exquisitely detailed and prompt response! I understand many things now. The fill choice is good then, because it will make it way easier then as I could select just 12 cards (one for each interval type) and make a lot of the work automated which is always great. Now, I will try to clarify points one by one where needed, and sorry for not being specific enough in some cases.

@Eltaurus

I’m not sure, what you mean by “single-choice questions”.

Well apparently single choice questions means that just a single choice is the correct answer (such as a radio button), which is what I need. Multiple choice questions can have more than one correct answer (so more of a checkbox). I just learned this a while back while researching up on how to do all of this, so its totally OK if you were also confused.

@Eltaurus

I’m not entirely sure what you are trying to achieve with this setup, though, and whether Cloze Cards are really suitable for your purpose. If you provide more details on your logic behind this, I might be able to give better explanations.

So basically, I have several piano diagrams that display two keys pressed (highlighted in blue) at once and that is posed as the question with audio being played. Then I have all of the musical intervals and an accompanying image with supplementary info about that interval as correct answers: minor second, major second, … all the way to an octave (12 in total). Of course I am interested in that my students are able to respond but just with the text part of that only, so corresponding supplementary answer images would only show when the answer is shown they don’t need to be part of selectable answers or tapping, though that would not be a bad idea because it would be like playing a match game. tapping on any of the words (minor, octave, ETC) or selecting an answer or typing can be a good way to review these, though thinking about ti a bit, typing probably is not as interactive or helpful for recalling this as its a much more visual approach.
Since there are just two intervals that have a single word as the correct answer, this is why I was asking if there needs to be something different if an answer must be two words to be correct.
Now, because many parts of the musical keyboard can produce any of these intervals more than once, the correct answers can be the same for 12 cards for each named musical interval. So there are 12 minor seconds, 12 major seconds and so on because they can be played in several different zones of the piano.

Point taken on review intervals as well, very sensible observation. And in regards to accessibility, I think putting these examples in documentation, I.E.: what you type versus what you get or how the programming responds when given certain input, in text helps the most, also describing screenshots or embedded images in docs is probably the most significant thing. Accessibility in anki still has some way to go, especially because some parts of the interface are not spoken correctly with screen readers but of course that is out of the scope of this template. Haven’t tested keyboard controls. Also I find that the audio button is not described correctly and I for some reason cannot focus it with keyboard from the windows app, though the native one that anki has does not present these problems.

[Eltaurus]

@ultraleetj

Well apparently single choice questions means that just a single choice is the correct answer (such as a radio button), which is what I need. Multiple choice questions can have more than one correct answer (so more of a checkbox).

Oh, I get it now. The template uses Memrise terminology, so all the multiple-choice cards are, in reality, what you are referring to as “single-choice”. Memrise (and consequently this template) doesn’t have actual multiple-choice questions, so it all should align with your purpose as is.

---

@ultraleetj

So basically, I have several piano diagrams that display two keys pressed (highlighted in blue) at once and that is posed as the question with audio being played. Then I have all of the musical intervals and an accompanying image with supplementary info about that interval as correct answers: minor second, major second, … all the way to an octave (12 in total).

Given that, I think regular Cards indeed will be a better fit. Cloze templates are suited for cases where the number of Cards generated from a single Anki Note can vary significantly depending on the content. In your case, the number of questions is fixed (and even the total number of Notes is predefined), so all of it can be hard-coded into the template. There are many ways to go about it, however, depending on your preferences.
The most straightforward, IMO, would be to simply make 12 different Anki Notes for each musical interval with the same name in the text field, and different images and sound files in the diagram and audio fields.
Alternatively, if you would prefer to keep all information on each musical interval on the same Anki Note, you could add more Fields, so that each Note would have a Field for the text name, 12 diagram Fields, and 12 respective audio Fields. The Card Types can then be duplicated to produce a separate Card for each of the images…
As a third alternative, you could place all the images in the same Field on the same Anki Note as a list, and get the Card to select a single random one as a question, while showing all associated diagrams on the back side at once. Or make Anki generate 12 separate Cards, each using the respective image from the list. The possibilities are practically endless here.

@ultraleetj

answer images would only show when the answer is shown they don’t need to be part of selectable answers or tapping, though that would not be a bad idea because it would be like playing a match game. tapping on any of the words (minor, octave, ETC) or selecting an answer or typing can be a good way to review these, though thinking about ti a bit, typing probably is not as interactive or helpful for recalling this as its a much more visual approach.

The template can be set up to produce any number of tests in various directions as you’d like. For example, you can set it to make the following Cards for each Anki Note:

1. Question: Audio; Answer: Text Name [Multiple-Choice] (diagram shown on the back)
2. Question: Audio; Answer: Text Name [Typing] (diagram shown on the back)
3. Question: Audio; Answer: Diagram [Multiple-Choice] (text name is shown on the back)
4. Question: Diagram; Answer: Text Name [Multiple-Choice] (audio is played on the back)
5. Question: Diagram; Answer: Text Name [Typing] (audio is played on the back)
   …

There are detailed instructions for these kinds of adjustments in the customization section of the documentation:

GitHub - Eltaurus-Lt/Anki-Card-Templates: Anki Card templates and the managing Add-on

Although they are still WIP and can get quite involved, if you have many changes to make.

Alternatively, there is a GUI for setting all of this up that is being developed. I’m not sure how much it will make things more accessible to you, but basically, it is a dedicated Anki window that allows specifying different Card Types in the format I listed above, using an interactive table:

If you come up with the setup you like in the format I specified in the example, I’ll be able to easily make it into a Note Type for you. Or, if you wish to experiment with it yourself and be able to make other Note Types in the future, I can provide you with the installation instructions.

---

@ultraleetj

selectable answers or tapping

Just to be clear, tapping is the mode that provides users with all the words the correct answer consists of and expects them to put the words in the correct order. So, if the correct answer on your Card is “minor second”, it will have only two buttons: “minor” and “second” (in random order), and with them the users will be able to input “minor second” or “second minor”. I still don’t think it makes sense to use this input mode for your use case.

---

@ultraleetj

I think putting these examples in documentation, I.E.: what you type versus what you get or how the programming responds when given certain input, in text helps the most

There is a brief description of different Field roles in this post: Memrise card template [support thread] - #75 by Eltaurus
I am planning to add more such explanations to the customization section I referenced above, so I’ll keep your advice in mind and try to make them more explicit.

@ultraleetj

describing screenshots or embedded images in docs is probably the most significant thing

I’m not sure how much I can improve on this one. The screenshots, the way I tend to use them, are mostly there to directly represent what is described right above them, so I rarely find anything meaningful to add to the alt text.

@ultraleetj

I find that the audio button is not described correctly

In what way? I’m not aware of any bugs associated with it.

[ultraleetj]

[quote=“Eltaurus, post:218, topic:34233”]
The most straightforward, IMO, would be to simply make 12 different Anki Notes for each musical interval with the same name in the text field, and different images and sound files in the diagram and audio fields.[/quote]
So this is referring to specifically creating each interval as a note type in Anki but using or cloning a note type from the template, or from scratch?

@Eltaurus

Alternatively, if you would prefer to keep all information on each musical interval on the same Anki Note, you could add more Fields, so that each Note would have a Field for the text name, 12 diagram Fields, and 12 respective audio Fields. The Card Types can then be duplicated to produce a separate Card for each of the images…
As a third alternative, you could place all the images in the same Field on the same Anki Note as a list, and get the Card to select a single random one as a question, while showing all associated diagrams on the back side at once. Or make Anki generate 12 separate Cards, each using the respective image from the list. The possibilities are practically endless here.

that third alternative sounds like the most automated one. TO be clear, there would be 12 different images for the question (as diagrams) and then there is a single image for the answer (that expand or complements information on the interval type of which there are also 12) so I would need 144 cards generated, 144 of these are unique diagram question images, and the answers would be repeated every 12 times for each interval.

@Eltaurus

The template can be set up to produce any number of tests in various directions as you’d like. For example, you can set it to make the following Cards for each Anki Note:

1. Question: Audio; Answer: Text Name [Multiple-Choice] (diagram shown on the back)
2. Question: Audio; Answer: Text Name [Typing] (diagram shown on the back)
3. Question: Audio; Answer: Diagram [Multiple-Choice] (text name is shown on the back)
4. Question: Diagram; Answer: Text Name [Multiple-Choice] (audio is played on the back)
5. Question: Diagram; Answer: Text Name [Typing] (audio is played on the back)
   …

Options 3, 4, and 5 would be exactly the ones i would want here, though audio would be part of the question in all of them as well.

@Eltaurus

Alternatively, there is a GUI for setting all of this up that is being developed. I’m not sure how much it will make things more accessible to you, but basically, it is a dedicated Anki window that allows specifying different Card Types in the format I listed above, using an interactive table

Could probably try to use it when its ready and give you some feedback.

@Eltaurus

So, if the correct answer on your Card is “minor second”, it will have only two buttons: “minor” and “second” (in random order), and with them the users will be able to input “minor second” or “second minor”. I still don’t think it makes sense to use this input mode for your use case.

Well, it depends, really doesn’t happen often, but I have had students definitely mixing up the order of these but yeah its not really practical in this case.

@Eltaurus

In what way? I’m not aware of any bugs associated with it.

Well, when I look for the audio button to play using this template, my screen reader reports a link as "by Lτ " which takes me to the git page for this template. I tab again and there is a “flip” button and then tabbing again there is a link without any text, screen reader just says “link” and that is the audio button. Though I think this appears to be also true for the Anki one as well.

[Eltaurus]

@ultraleetj

So this is referring to specifically creating each interval as a note type in Anki but using or cloning a note type from the template, or from scratch?

Note type is a template for Notes – you will only need one for all Cards you want to make here. After the Note Type is set up, you can use it to create multiple Notes – all will have the same sets of Fields (filled with different content) and will generate same Types of Cards as defined by the Note Type settings. That first scenario I described suggests using 144 Notes (12 Notes per interval). You can use Note-cloning Add-ons or sticky Fields to avoid any unnecessary manual work.

@ultraleetj

that third alternative sounds like the most automated one.

Just keep in mind that automation itself takes extra work to set up, and for as few as about a hundred Cards, it might be not worth the effort. It would be a different thing if you had thousands of Cards, or needed to be able to easily make more Cards from this template in the future.

@ultraleetj

there would be 12 different images for the question (as diagrams) and then there is a single image for the answer (that expand or complements information on the interval type of which there are also 12)

I don’t think you’ve mentioned it before, so I was under the impression that the answers would be text names for the intervals. What kind of images are supposed to be the answers, if the diagram images are the questions?

@ultraleetj

I would need 144 cards generated, 144 of these are unique diagram question images, and the answers would be repeated every 12 times for each interval.

Also, to clarify things, all the scenarios I’ve described previously involve generating 144 Cards (actually 144 multiplied by the number of test/directions per each musical interval instance, but I’ll omit it for simplicity here). The difference is in how the information for those Cards will be distributed between Notes. In the first scenario, there will be 144 Notes, each of which will contain only the necessary information for a single question and generate a single Card. In the second scenario, there will be 12 Notes, each containing complete information on all 12 instances of the same musical interval (distributed over 12 separate Note Fields) and, respectively, generating 12 different Anki Cards. In the third scenario, there will be the same 12 Notes with 12 Cards per Note, but the related information on each Note will be gathered inside the same few Note Fields. The last one might be more convenient to manage in Anki Editor, but will be trickier when it comes to multiple-choice questions.

@ultraleetj

Options 3, 4, and 5 would be exactly the ones i would want here, though audio would be part of the question in all of them as well.

This can be done. Although now that you’ve mentioned the answer being an image as well, I’m again not quite clear on the complete setup. Could you maybe share and example of your content? And are there any other types of testing directions/types you’d like to add to the three mentioned here?

@ultraleetj

Could probably try to use it when its ready and give you some feedback.

It is ready in terms of being fully functioning as of now. It’s just that testing and documenting things takes a lot of time as well, so I probably won’t get to uploading this version to AnkiWeb for quite a while. You can already install it from GitHub, though. It will be more laborious than installing an add-on from AnkiWeb, but not by too much. I can detail all the steps if you’d like.

@ultraleetj

Well, it depends, really doesn’t happen often, but I have had students definitely mixing up the order of these but yeah its not really practical in this case.

If it’s just a few cases, you can simply manually add the wrong-order combinations to the list of choices for multiple-choice Cards.

@ultraleetj

Though I think this appears to be also true for the Anki one as well.

It most likely does. On the desktop app, the template uses the regular Anki buttons, without altering any of their code. Only the appearance is adjusted to fit the Memrise style.
You can try adding the following to the “user posterior scripts” section on the front template:

document.querySelectorAll("a.soundLink.replay-button").forEach(L=>L.setAttribute('aria-label', 'audio button'));

(if you are planning to use it on AnkiDroid, the same should be done on the back template as well). This should make all the audio buttons be read as “Link: Audio button”, if that helps.

[ultraleetj]

@Eltaurus

I don’t think you’ve mentioned it before, so I was under the impression that the answers would be text names for the intervals. What kind of images are supposed to be the answers, if the diagram images are the questions?
Could you maybe share and example of your content? And are there any other types of testing directions/types you’d like to add to the three mentioned here?

sure, though forgive if this dimensions things funny… have no way to check if it looks OK. The question is a keyboard diagram that will play a sound when shown:

and the answer or back of card is yeah, well the text name for the interval, plus an image that is designed to just complement that info, a simple visual representation of it, as in the following:

as for testing directions… not sure if they would be necessary, for example to reverse the cards because the idea is that the visual/audible recognition of the interval is what is being tested.

@Eltaurus

It is ready in terms of being fully functioning as of now. It’s just that testing and documenting things takes a lot of time as well, so I probably won’t get to uploading this version to AnkiWeb for quite a while. You can already install it from GitHub, though. It will be more laborious than installing an add-on from AnkiWeb, but not by too much. I can detail all the steps if you’d like.

Point me to it and I will send some feedback with my first impression of the interface navigation ETC.

@Eltaurus

If it’s just a few cases, you can simply manually add the wrong-order combinations to the list of choices for multiple-choice Cards.

yeah this could work.

@Eltaurus

You can try adding the following to the “user posterior scripts” section on the front template:..
This should make all the audio buttons be read as “Link: Audio button”, if that helps.

Yes, this was spot on. Worked perfectly. Thanks a lot!

[Eltaurus]

@ultraleetj

the answer or back of card is yeah, well the text name for the interval, plus an image that is designed to just complement that info, a simple visual representation of it, as in the following:

I think I get it: that image itself is not actually an answer, but a kind of mnemonic to be shown on the back of a Card after the answer is submitted by a user. Akin to what the “Extra” Field in the default setup is intended for.

So, as I understand so far, the setup you are looking for would involve a Note Type with 4 Fields (Text name, Diagram, Audio, Mnemonic) and 3 Card Types I described previously (plus 2 choices Fields for the two kind of multiple-choice Cards, but you won’t need to make those yourself if you’ll be creating the Note Type with the GUI). There will be 12 groups of 12 Notes for each interval (with the Notes inside the same group having the same Text name and, I suppose, Mnemonic, while containing different Diagrams and Audio files), totalling at 12*12=144 Notes. Each Note will be generating three different Types of Cards, yielding 144*3=432 Cards overall.

@ultraleetj

Point me to it and I will send some feedback with my first impression of the interface navigation ETC.

1. To avoid conflicts, you’ll need to disable your currently installed version of the template support add-on:
   go to Tools → Add-ons → Select the Memrise Cards Lt in the list → click Toggle Enabled button (you’ll be able turn it back on at any point in the future the same way)
2. In the same window, deselect the add-on and click the View Files button to open the general Anki “addons21” folder. This is where all manually installed add-ons should be copied
3. On GitHub, go to the following branch: GitHub - Eltaurus-Lt/Anki-Card-Templates at Note-Type-Creator
4. Navigate to the Code button (read as “Code. Button. Collapsed. Has pop-up.”)
5. From the pop-up menu, select Download ZIP
6. Extract the downloaded archive and move the “Add-on” folder from it directly into the “addons21” folder from step 2
7. Restart Anki

To make the Cards using this version:

1. Go to Tools → New Memrise (Lτ) Note Type to open the Note Type Creator window
2. Fill in the name for the Note Type. Add, delete, or rename Fields and Card Types to replicate the desired setup.
3. (Optional) Type a new name into the preset text line and click Save to be able to make adjustments to the setup later (already created Note Types can’t be edited using this window, but the saved presets could be modified and used to make new similar Note Types)
4. Click Create
5. Use the created Note Type to make the Notes as usual: click Anki’s Add button in the top center menu → select the Note Type in the “Type” menu (the name you used during the creation will be prefixed by "Memrise (Lτ) ") → Fill in the Fields with the respective content → toggle Sticky on the text name (and mnemonic) Field to keep its content the same for several subsequent Notes → click Add
6. Select all the created Notes in Anki browser and use on them the “Fill Choices” function to populate text name choices. In the dialog window, the Source Field should be set to what you name the text name Field itself (e.g., “Interval”), and the Choices Field will be the same, prefixed by “Choices” (i.e., “Choices Interval”).
7. Use “Fill Choices” a second time to populate Diagram choices in the same way

[ultraleetj]

@Eltaurus

So, as I understand so far, the setup you are looking for would involve a Note Type with 4 Fields (Text name, Diagram, Audio, Mnemonic) and 3 Card Types I described previously (plus 2 choices Fields for the two kind of multiple-choice Cards, but you won’t need to make those yourself if you’ll be creating the Note Type with the GUI). There will be 12 groups of 12 Notes for each interval (with the Notes inside the same group having the same Text name and, I suppose, Mnemonic, while containing different Diagrams and Audio files), totalling at 12*12=144 Notes. Each Note will be generating three different Types of Cards, yielding 144*3=432 Cards overall.

Correct. But then how many choices would be displayed in this case? 4 or more could be good.

@Eltaurus

2. Fill in the name for the Note Type. Add, delete, or rename Fields and Card Types to replicate the desired setup.

Unfortunately, my success stops here. Installing the add on was straightforward, and I was able to open the window for it. However, the note type name, the first edit field reads a lot of things, I guess to do some kind of visual demonstration, maybe they could be filtered out somehow, because at the time my screen reader says
"edit< nobr>Name for the created Note Type < /nobr >
< i>e.g. < /i> “Greek”, “History”, < nobr >“Spanish (no audio)”, or < / nobr><br

“Geography (multiple-choice)”"

Basically reading the entire markdown code with all signs. Then after a bit I heard editable, selected, New Note Type.. and that is how i knew it was the one.
I tab and then get a color note for all types. This reads and functions correctly.
Same for the preset, combo box, though some html tags get also pronounced.

However, as soon as I enter the fields area and tab past the first field, the table is presented and some headers, things are read, like math, big… but have no idea what these are or do. I tried navigating it like a regular table, but then the cursor jumped all the way to the extra 2 field. Also, using hierarchical navigation in elements shows three (unlabelled) checkboxes and two unlabelled edit boxes for each field before finding the delete button. And then I cannot get focus out of that table. I also found the card types area but was not able to get propper keyboard focus to them, though navigating and manipulating the values is fine and I think I know how these work by intuition, though also cannot get focus out of that area, so am pretty sure this is really a QT 6 bug.
(edit)
I think I am getting the hang of this. For fields: Its a table with 7 columns and the rows depend on the number of fields you make. The checkboxes are not unlabeled, I guess they are the middle three values: big, back, math, which not sure what they do, but as with QT, its not really the best interface so takes some figuring out. Then the two “unlabelled” edit boxes are static keys and random keys? and a last column which is read as “data element” is actually the delete button. In fact, all columns and headers read like data elements.
For card types: Navigation works perfectly here using standard screen reader table navigation commands. I guess you select all using combo boxes and there is an edit field with a prompt I.e. (type what you hear) that gets shown to the user? this one is definitely much more straightforward to understand and do. However, is there a “back extra” and not front extra possibility?

Now knowing all of this I could throw a paragraph or two on how to navigate/use this window in docs for the editor you made.

I guess adding a few keyboard shortcuts to focus specific sections can get around this limitation. It is however, doable by using more convoluted navigation tricks. Ii will post back with my progress. Thanks a lot.

[Eltaurus]

@ultraleetj

how many choices would be displayed in this case? 4 or more could be good.

It will be 6 by default. You can adjust this by putting

mchOptionsN = 10;

into the front “user prior scripts” section on the respective Card (change the 10 to the desired number)

@ultraleetj

Its a table with 7 columns and the rows depend on the number of fields you make. The checkboxes are not unlabeled, I guess they are the middle three values: big, back, math, which not sure what they do, but as with QT, its not really the best interface so takes some figureing out. Then the two “unlabelled” edit boxes are static keys and random keys? and a last column which is read as “data element” is actually the delete button. In fact, all columns read like data elements.

Yes, all of this is correct. The column headers should have tooltips that are supposed to explain what each of them does. I guess they are not being read to you (despite tooltips on other elements being read as intended), so I’ll have to look into this.

However, in the “Fields” table, you don’t need to change anything other than the number of Fields and their names. You already identified the buttons for deleting each of the Fields in the last column (the button for adding a new Field is located outside of the table, and seems to be selected right before it, at least when I cycle through the elements, for some reason). And renaming is done in the First column (the “Extra 2” you encountered is the name of one of the Fields in the default preset).

As for the “Card Types” table, it will require more set up. It has the following 8 columns:

1. Card Type name (a simple descriptor like “Listening” or “Naming”, etc.)
2. Dropdown menu for selecting the Question Field (the list items correspond to the Fields set up in the “Fields” table)
3. Similar dropdown menu for the Answer Field
4. Input method (dropdown to select either “Typing”, “Multiple-Choice”, or “Tapping”)
5. Text prompt that is supposed to instruct a user on what to do with this type of Card when it is reviewed (it is kind of a Memrise quirk; as the instructions are usually self-evident from the context, this Field can be set up to empty)
6. A “Front Extra” Field. Also a Memrise quirk, I would suggest set all to blank (ー) for your case
7. A button to delete the Card Type
8. A button to clone the Card Type (adds a new Card Type as the bottom row of the table, with all the cells inheriting their content from the progenitor Card Type)

@ultraleetj

Now knowing all of this I could throw a paragraph or two on how to navigate/use this window in docs for the editor you made.

This will be appreciated.

@ultraleetj

I guess adding a few keyboard shortcuts to focus specific sections can get around this limitation.

That’s a good idea. I like keyboard shortcuts in general, but I’ll need to think about how to better set them for the Qt window. I also took a note to add explicit labeling to the checkboxes and buttons in the table, which will require some research on my part too.

@ultraleetj

Thanks a lot.

You are welcome. And thank you as well, the feedback you provided helps quite a bit.

[ultraleetj]

@Eltaurus

2. Dropdown menu for selecting the Question Field (the list items correspond to the Fields set up in the “Fields” table)
3. Similar dropdown menu for the Answer Field
4. Input method (dropdown to select either “Typing”, “Multiple-Choice”, or “Tapping”)
5. Text prompt that is supposed to instruct a user on what to do with this type of Card when it is reviewed (it is kind of a Memrise quirk; as the instructions are usually self-evident from the context, this Field can be set up to empty)
6. A “Front Extra” Field. Also a Memrise quirk, I would suggest set all to blank (ー) for your case
   the cells inheriting their content from the progenitor Card Type)

OK. SO had a go at this and was able to create the 3 card types:

1. Multiple choice from answer texts
2. Multiple choice using answer images (mnemonics)
3. Making the user write the answer texts.

also for fields I had:

1. answer text
2. question diagram
3. audio
4. answer image

However, I am not sure how can I have the audio show together with both the question diagram, and the answer text and image? Tried modifying the code in the card templates directly but I broke something. Also, I would like the answer image to show on the back and not on front of card.
I am using the prompts for the input as well, because it just adds a bit of “formality” or guidance.
I also figured I have to do the written cards first, then the other ones with the choices I guess it would be easier to just duplicate all of these and then just change the template that is being used…