中文 Literacy Speedrun II: Character Cyclotron
blog.kevinzwu.com
blog.kevinzwu.com
Edit addendum: https://blog.kevinzwu.com/chinese-cursed-logographic-dags/ is a fun read. I've been using the imaginatively titled "kanji study" app, which uses the same Outlier database mentioned which has the graph based etymology.
There's an additional level of chaos when learning the "same" characters as kanji rather than hanzi.
Also for anyone who speaks or is currently learning Chinese... I've been working on a multiplayer CJK word game that shares a similar efficient brute force style of learning to the author's approach (although presented via gameplay instead of tooling). Every turn you get a random character and must type in a word that contains the char in ANY position. If you like fast paced word games it might be up your alley: https://danobang.com/?game_lang=cmn
I remember learning about your game in a mandarin forum. Thanks for making it, it's very cool!
> I would then click on the component characters to open their nested dictionary entries. SLOW!
> If I needed to remember the stroke order, I would scroll down for the static display. SLOW!
So, all of these are included in Anki-xiehanzi(https://github.com/krmanik/Anki-xiehanzi). Free open source software like Anki & xiehanzi can save you from using all those tokens.
It is also allowing me to see all relevant associations easily when revealing the card in built in SRS, you add cards to SRS as you browse, so they are related to what you already know / currently exploring.
Mind you, all data visible is collected from different reputable available sources. When you click "explain" there's a clearly marked LLM explanation, but my explanation generation pipeline pushed all generated explanations through 5 different models including all top Chinese-first for verification, and on average it took a few iterations back and forth to iron out any information that could potentially mislead the learner.
You can actually see thousands of words I typed just working on that pipeline here https://hanzirama.com/making-of
It's not opon source but it is completely free. Open sourcing is on the table but currently it would be additional work and distraction. Just licenses seem like a headache, from a quick poke even when ok for use on non-commercial data redistributing may not be ok. So not any time soon, but if you are working on something similar I can at least share detailed datasources, finding good ones was not easy and LLM integrate them fast.
What do you mean by that? Is this relating to the LLMs or the datasets used as inputs or something else?
0. https://claude.ai/share/23d27617-7608-4c58-93b9-b9613bc29318
Hm. I always knew voice mode was a thing, but I have never tried it. What's people's experience with it?
Being able to correct my words is a good thing. Hell, I did it ~3 times when writing this comment. I can't do that when I'm rambling. I'll trip, or CC will think I'm finished.
A shame that this amazing resource is not linked.
i’ve used this for a brief while, but dropped practicing handwriting completely shortly after.
WIP (need more work in multi-hanzi words), but won't stay in the same 5 words for more than a day. it has been working well for me
the most interesting thing was GPT helped with the sentences and simplified words meaning and bing translate provided the audios
the goal is get the ~2000 words you need to be proficient in 1 year, 5 words a day plus refresh old words, also it keep track of your progress against the year, no streaks
>Empty landing page with a "Sign in with Google" button
>Can't find anyone else talking about it online, no screenshots of the gameplay, nothing
that's exactly what it does. free. no ads.
yeah, google it. nobody in there, you won't enter, well you're not a maverick in curve of adoption. that's fine.
hand over your credentials? what's credentials? a Google oauth login?
uh?
...Why? That advice is universal for a reason. The side adventure with Claude Code strikes me as a distraction from the fact that there is a hard thing you want to do but are avoiding because it's hard.
Meanwhile others read books and get pretty good at their language of choice in a couple years.
2. Split into sentences and tokenize sentences into words, e.g. using https://github.com/fxsjy/jieba
3. Count how often each word appears and sort sentences by descending frequency of the least common word.
4. Use binary search to find a location in the sorted collection of sentences where the difficulty feels about right.
Of course this gives you a collection of disjointed sentences, but you can always go to the original file and look at the surrounding context when you find an interesting or confusing one.
My feeling is this level is just too early to read "real" texts, so I am continuing to just use graded readers. I use the Du Chinese app for this, it contains a bunch of short stories at different comprehension levels, and has a spoken accompaniment to each story read by a real speaker (not AI/TTS). I also have some physical books from LingLing Mandarin, I like the challenge of not having a dictionary immediately to hand like I do in the app. My hope is by the time I finish with the Advanced stages of each of these sets of readers, I will be able to start reading "real" texts and fill in gaps with a dictionary app, at which point there is an infinite supply of material.
I do worry I'll end up at the "10% missing comprehension" described in the article, though, at which point I guess I'll try to find even higher level graded readers, if they exist. We'll see.
Ain't nobody downloading stuff from GitHub and splitting things into tokens to learn English like the other commenter is suggesting lmfao. But people are obsessed with bizarre and completely inhuman methods like that when they learn Asian languages. I cannot understand why, and it's seriously just Asian languages.
With a good order (RTK), optimal reviews (SRS) and putting 30 minutes a day it is possible to learn keyword to Kanji writing in a couple of months to one year. Make it two if you are a busy person. After that you need 10 minutes to maintain the knowledge per day. (I’m assuming 2200 Kanji).
People that did that successfully will recommend it to be done as early as possible as they know the boost in learning it provides.
I think it’s a trap, because it’s possible to get to a very useful level in the language while ignoring Kanji, and most people will be perfectly happy staying there. At that point you will have a much better idea if you really need to go all the way.
It’s like thinking you can get good at chess by just memorizing how each piece moves. You need the board, the surrounding "context", and not just study in isolation.
Under normal circumstances, even if you grind out "knowing" all the words in advance you will still struggle to read any basic sentence and you've essentially wasted time because it's an unskippable step; to be good at reading you need to read a lot. He seems to already know Chinese though, so this might work for him since he is not really having to learn the language or specific vocabulary, just how it's actually written.