I’m a developer and I’ve been thinking for a while about a feature for getting a list of the most commonly occurring words per episode of what you’re watching!
I think this is a great way to optimize the language immersion process. By creating Anki cards for these top occurring words, you’ll be able to increase your general comprehension for a given show quicker, and as a result increase your opportunities for comprehensible input and language acquisition. I believe this would be trivial to implement in the long run and exponentially useful for users.
The biggest edge case would be filtering out the most common words that appear in any sentence, like the or a. I think a good brute force solution is to select only the words above your current vocabulary highlighting level.
For example, since I am level 5: 4000 in Japanese, I render a list of n items, lets say 10, that are ranked after the most commonly occurring 4000 in the language, but are the top occurring in the current episode.
A naive ranked list of words is also fine, because a user can reference the list and look for good sentences to make Anki cards from while immersing in the language.