Converting a language learning book to Anki

I have a French language course book + CD.
My goal is to scan the “French phrase left and German phrase right” tables and the French left and German right vocabulary tables and automatically convert them to Anki so I don’t have to type them in, I just tell Anki: I finished chapter 3, add all the new words and phrases. I also want the very charming sounding pronunciation from the CD.

Of course, I could do this manually, using Audacity and so on.

So my aim is not only to recognise the table automatically and extract into a format for anki, but also to force align the text with the audio so that I can extract the audio for the phrase/word, you get the idea.

Sounds like a lot of tinkering, so I wonder if this has already been done.

PS: I’m aware of TTS, but that doesn’t solve the fact that I have to type in all the vocabulary, and that I’m probably going to get the wrong pronunciation with the robotic TTS voice.

2 Likes

I’d imagine you need to do this by yourself, but there are ways that you can make it easier for yourself.
For the text, maybe you can use Tesseract OCR and parse the tables into spreadsheets with Pandas? Then for the audio, use OpenAI’s Whisper to find out where to split the track for each word. All of that can be done in Python.
Then when you have the book&CD parser done, you can use the data from that to feed into Anki decks.

3 Likes

Sounds like a plan, but I am waiting for someone else to chime in with hopefully a more automatic process already established.

I do this regularly.

I copy or type in the exercises and then their model responses. I set them up for Reverse capability as well.

I do something similar with the vocabulary. Usually, I do the vocabulary cards first. If they are useful, I then create the exercise cards.

Lastly, I create my own cards that test me for conjugation and declension; sometimes using cloze, if that makes sense, otherwise building sentences that require e.g. the preterite tense or the dative case.

Takes a little time, but you must learn and get an idea at least of what you want to remember, before you can profit from Anki.

In that way, you have effectively turned the book into an Anki deck or a collection of Anki decks. You have also learnt some of the intricacies of the language in producing those decks.

3 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.