Hey everyone!
I’m trying to get back to reading visual novels on PC but all of the resources that I have found are either 1) no longer maintained 2) severely outdated, or 3) no longer supported (RIP Yomichan).
What’s the most recently maintained text extractor these days? I got Yomichan’s successor, Yomitan, but haven’t found anything new so far as to text hookers. Any guidance appreciated! 
I don’t bother to text hook as often as I used to, so maybe there’s new stuff I don’t know, but I’m still using textractor most of the time which is kinda fiddly like it always was but works out a lot of the time. Agent is worth a look too. Anything it works on it works instantly with no fuss, but because it has to pull from scripts people make for it its library is limited to a specific, set list of compatible VNs.
Oh thanks! From what I read, people made it seem like textractor was borderline unusable lmao but if it’s just a little janky, then I don’t mind! I’ll try and download it again. 
It’s gotten me through lots of VNs! You’ll definitely find things it’s just not gonna work on depending on the engine the VN is on, unfortunately, but there’s a lot it does too. You really just have to point it at the program, advance a line or two, and then you start looking through the drop down it generates to see if any option there has the correct text. If it’s gonna work, that’s often good enough. There’s an option to search for the text on screen but I’ve personally had limited success getting that to work out. Sometimes people will also have a hook code, a long string of characters they’ve already figured out to point textractor to the exact right place. Might not run into that that often, but it’s worth keeping in mind as an option anyway; it’s the only way I’ve been able to do a few things.
I failed to make work it on linux but agent seems to be good: GitHub - 0xDC00/agent: Universal script based text hooker (powered by FRIDA).
There’s also the https://sugoitoolkit.com/ which does seems to be something interesting too OCR wise. Again, I did not experiment firsthand.
Thanks! Texthooker doesn’t seem to work on games run on emulator, so I’ll try this.
GitHub - matt-m-o/YomiNinja: Open-source OCR and dictionary tool. is OCR that works on almost anything. None of the tedious text hooker fiddling required
Woah, thank you! This one’s great!
Update: I successfully set this up. It’s better than what I had in mind. So quick. Thanks again!
Game Sentence Miner is currently the most comprehensive(?) OCR/webhook mining tool out right now. You can mine the audio directly from the game without much hassle at all (it does all the trimming and editing behind the scenes), and makes use of Yomitan (Yomichan’s successor) so you can just import your dictionaries over. Also has an overlay feature so you can do lookups right on the text of the game.
Agent is the current texhooker for modern-ish games (emulation too), while Textractor (chenx221 fork) is for VNs
TIL I can probably make Subs2srs decks out of video games, if I really wanted to. (I have enough shadowing material lol, but good to know)