[X] My Assistant

I've never translated manga pages in my life before, although I know hirigana, katakana and a couple of kanji. Is there a guide to do it or a normal efficient method? (IMG:[invalid] style_emoticons/default/ohmy.gif)

I like to first have a unicode script in MS Word or something so I can Google search phrases or kanji that I don't get but I've never tried to type the characters before, so I've just been using OCR to copy and paste the characters, double-checking that there aren't any mistakes and adding any kanji that it misses. For manga style though, ABBYY almost never autodetects text so I have to go in manually and draw the boxes which partially defeats the purpose. I started off with pen and paper, but if I had issues with something (almost only kanji :V) I can look it up quickly without breaking flow if I have the characters already in unicode text format.

Is there a better method for this? Also, is there a place I can go to view stylized fonts? Japanese characters originated from calligraphy which, a stylized version is basically the same as English is to cursive writing in that some characters look nothing alike. It gets really hard to figure out what some of them are.

This post has been edited by BeachedFail: Feb 18 2018, 00:40

I've never translated anything in my life before. Is there a guide to do it or a normal efficient method? (IMG:[invalid] style_emoticons/default/ohmy.gif)

I like to first have a unicode script in MS Word or something so I can Google search phrases or kanji that I don't get but I've never tried to type the characters before, so I've just been using OCR to copy and paste the characters, double-checking that there aren't any mistakes and adding any kanji that it misses. For manga style though, ABBYY almost never autodetects text so I have to go in manually and draw the boxes which partially defeats the purpose. I started off with pen and paper, but if I had issues with something (almost only kanji :V) I can look it up quickly without breaking flow if I have the characters already in unicode text format.

Is there a better method for this? Also, is there a place I can go to view stylized fonts? Japanese characters originated from calligraphy which, a stylized version is basically the same as English is to cursive writing in that some characters look nothing alike. It gets really hard to figure out what some of them are.

I'm pretty old-fashioned. I just read the Japanese source text and directly translate it into English, look it up in a paper dictionary, or if need be just type it in and use it to search online. OCR sounds interesting, but I've never used or needed it.

As for stylized fonts, all i can say is that it gets easier with experience. Read as much material, of every sort, in Japanese, and expose yourself to all of the different fonts and styles out there. You'll pick it up in time.

I've never translated manga pages in my life before, although I know hirigana, katakana and a couple of kanji. Is there a guide to do it or a normal efficient method? (IMG:[invalid] style_emoticons/default/ohmy.gif)

I like to first have a unicode script in MS Word or something so I can Google search phrases or kanji that I don't get but I've never tried to type the characters before, so I've just been using OCR to copy and paste the characters, double-checking that there aren't any mistakes and adding any kanji that it misses. For manga style though, ABBYY almost never autodetects text so I have to go in manually and draw the boxes which partially defeats the purpose. I started off with pen and paper, but if I had issues with something (almost only kanji :V) I can look it up quickly without breaking flow if I have the characters already in unicode text format.

Is there a better method for this? Also, is there a place I can go to view stylized fonts? Japanese characters originated from calligraphy which, a stylized version is basically the same as English is to cursive writing in that some characters look nothing alike. It gets really hard to figure out what some of them are.

For someone new to translation, I'd recommend you transcribe everything into a text file. It adds time the overall process initially, but usually saves time later. It's important for various reasons.

• If an unknown kanji comes up once, it'll likely come up again. Having it available to copy/paste saves a lot of time later.
• If you don't understand something you can copy/paste the whole line easily.
• It makes it easier for editors to see that they're putting your translation into the correct bubbles.
• It allows for easy use of CAT (Computer Assisted Translation) tools. Note that CAT is very different from machine translation. They're specialized programs that help to improve the workflow of translators.
• It allows you to separate the tasks of reading and translating. Sometimes my brain gets too fried to translate, but I can still mindlessly slog through and type out the lines and translate them later.
• When translating sound effects, it allows you to search/replace en masse. I usually try to add some variety to things like *thump*/*thwack*/*ba-dong* but for things like *sigh* there's really only one way to translate them all.

I'd strongly recommend against using OCR. For one, it's a crutch that will hold you back and keep you from developing your reading skills. For another, I've never seen an OCR tool that didn't make mistakes. When they do, they're usually very hard to spot because your brain wants to see what it expects to see, not what's actually on the page.

タ夕
りリ
カ力
持待
未末

Can you tell the difference between these pairs? These are just a small example of what I've seen OCR programs mix. If you typed them out by hand, you'd get the correct character. But if you OCR it and try to paste it into a dictionary, well... try it with these and see what happens! (IMG:[invalid] style_emoticons/default/tongue.gif) You'll see that these have very different meanings.

For reading stylized fonts, that's something that gets better with time and practice. Just try to read as much as you can and you'll get better and better as time goes on.

Beyond that, practice makes perfect!

OCR as in online character reader? I tried one of those some time ago and the results were full of errors, completely unsuit for a translation. I'd recomend a dictionary site like jisho.org which lets you look for kanji by radicals (or tagaini jisho if you need an offline program). A physical dictionary should do as well if you find it.

If you're new to this and don't know where to start I recomend:
1) Learn hiragana and katakana (or just keep the tables around so you can look them up and learn them while you read, be sure you understand how Digraphs and Diacritics work, it's not complicated).

2) Be sure you understand grammar and common expressions. A lot of the meaning of a Japanese sentence comes from little variations, any basic grammar book should work.

3) Use the dictionary whenever you find a word or a kanji you don't know.

4) Expand your kanji knowledge. This is not necessary if you only occasionally translate something but it makes the job quicker than using a dictionary for each word. You can do this systematically by learning kanji tables as they are tought in school or just trying to memorize whatever you encouter while reading manga.

tip: using microsoft ime (or the equivalents on linux/mac) to write in japanese can make it faster to look for words formed by kanji you already know or to write in hiragana/katakana.

This post has been edited by progste: Feb 18 2018, 16:40

OCR as in online character reader?

I may be being a bit nitpicky here, but OCR usually stands for "Optical Character Recognition".

For someone new to translation, I'd recommend you transcribe everything into a text file. It adds time the overall process initially, but usually saves time later. It's important for various reasons.

• If an unknown kanji comes up once, it'll likely come up again. Having it available to copy/paste saves a lot of time later.
• If you don't understand something you can copy/paste the whole line easily.
• It makes it easier for editors to see that they're putting your translation into the correct bubbles.
• It allows for easy use of CAT (Computer Assisted Translation) tools. Note that CAT is very different from machine translation. They're specialized programs that help to improve the workflow of translators.
• It allows you to separate the tasks of reading and translating. Sometimes my brain gets too fried to translate, but I can still mindlessly slog through and type out the lines and translate them later.
• When translating sound effects, it allows you to search/replace en masse. I usually try to add some variety to things like *thump*/*thwack*/*ba-dong* but for things like *sigh* there's really only one way to translate them all.

I'd strongly recommend against using OCR. For one, it's a crutch that will hold you back and keep you from developing your reading skills. For another, I've never seen an OCR tool that didn't make mistakes. When they do, they're usually very hard to spot because your brain wants to see what it expects to see, not what's actually on the page.

タ夕
りリ
カ力
持待
未末

Can you tell the difference between these pairs? These are just a small example of what I've seen OCR programs mix. If you typed them out by hand, you'd get the correct character. But if you OCR it and try to paste it into a dictionary, well... try it with these and see what happens! (IMG:[invalid] style_emoticons/default/tongue.gif) You'll see that these have very different meanings.

For reading stylized fonts, that's something that gets better with time and practice. Just try to read as much as you can and you'll get better and better as time goes on.

Beyond that, practice makes perfect!

Yeah, I'm starting to realize OCR isn't actually speeding me up all that much. I finally figured out how to swap my keyboard on the fly from romaji style Japanese input and English input which has sped me up considerably in the transcribing process, and since then OCR hasn't been too much of a benefit outside of kanji characters. While yes, most of the kanji are first or second reading level, that's still about 250 kanji. :<

I definitively did spot it making mistakes from time to time, usually with the Kanji so I always proofread it after it did it's thing. I didn't know about CAT though, which sounds like it will help immensely.

OCR as in online character reader? I tried one of those some time ago and the results were full of errors, completely unsuit for a translation. I'd recomend a dictionary site like jisho.org which lets you look for kanji by radicals (or tagaini jisho if you need an offline program). A physical dictionary should do as well if you find it.

If you're new to this and don't know where to start I recomend:
1) Learn hiragana and katakana (or just keep the tables around so you can look them up and learn them while you read, be sure you understand how Digraphs and Diacritics work, it's not complicated).

2) Be sure you understand grammar and common expressions. A lot of the meaning of a Japanese sentence comes from little variations, any basic grammar book should work.

3) Use the dictionary whenever you find a word or a kanji you don't know.

4) Expand your kanji knowledge. This is not necessary if you only occasionally translate something but it makes the job quicker than using a dictionary for each word. You can do this systematically by learning kanji tables as they are tought in school or just trying to memorize whatever you encouter while reading manga.

tip: using microsoft ime (or the equivalents on linux/mac) to write in japanese can make it faster to look for words formed by kanji you already know or to write in hiragana/katakana.

I haven't had too many errors with OCR in part because the manga pages are high resolution, but yeah it still slows me down in the transcribing enough that I'm not going to use it anymore.

I do know most of the hirigana and katakana characters, although I did find a pair of charts on textfugu.com that work really well for reference. Kanji grind me to a screeching halt though (IMG:[invalid] style_emoticons/default/anime_cry.gif) .

Is there a good place to read about grammar, and specifically how they use certain characters? The manga I'm translating tends to use a lot of "..." and long hyphens. My actual working knowledge of Japanese is basically a patchy selection of conversational Japanese so I tend to look up a lot of stuff during translation. Jisho will definitively be useful, thanks.

"..." and hyphens are just usually a pause, not really part of the grammar, but more of manga expressivity. Sometimes hyphens are used with kana for some english words, for example ほーりー (holy).
Also make sure you don't confuse it with the character ichi: 一
If you're in doubt just ask.
For grammar in general I used a series of books called Genki, but I'm sure you can find it in other places.
When in doubt on what some expressions mean it's usually enough to google them in romanji and you'll find some explaination.

This post has been edited by progste: Feb 18 2018, 22:38

Hey guys I need help with number 1 and 2 in this panel.

Hey guys I need help with number 1 and 2 in this panel.

It's more or less:
1- しかし下心がバレて嫌われたら元も子もない！あくまでさりげなく！
But if my ulterior motives are discovered and I become hated, (I'd lose everything / I'm done for)! I have to act casual to the very end!

2- 「なんとなく選んだらたまたま内容がエロかった」というギリギリひかれないラインを…！
"I just picked it out a random, and it just happened to be something perverted." I'll go with a (barely passable)* line like that...!
*ギリギリ is "barely/borderline/etc.", and ひかれない is the negative passive form of 引く, which in this case basically means "repulsive". So more literally, it'd be "barely non-repulsive", but that sounds weird. He's trying to have an excuse ready that keeps him from sounding too much like a pervert.

It's more or less:
1- しかし下心がバレて嫌われたら元も子もない！あくまでさりげなく！
But if my ulterior motives are discovered and I become hated, (I'd lose everything / I'm done for)! I have to act casual to the very end!

2- 「なんとなく選んだらたまたま内容がエロかった」というギリギリひかれないラインを…！
"I just picked it out a random, and it just happened to be something perverted." I'll go with a (barely passable)* line like that...!
*ギリギリ is "barely/borderline/etc.", and ひかれない is the negative passive form of 引く, which in this case basically means "repulsive". So more literally, it'd be "barely non-repulsive", but that sounds weird. He's trying to have an excuse ready that keeps him from sounding too much like a pervert.

Thanks!

I am not too sure with number 2.

Need help here.

I am not too sure with number 2.

In 1, he (assumed) is talking about how the right movie can get her excited and in the mood for sex, but in 2, he says that it looks like today's mood will be like a wake, ie no chance of sex.

Need help here.

ファミスタもどき/famista-like (lesson)? I'm not sure it will fit with the story but I may figure it out if I see the source.

艶技
sex techniques? 艶's nuance depends on what they'll actually do in the lesson.

Need help translating this! From Kancolle's doujin

キラ付けくらいちゃんと随伴艦付けてくださいよ!

What is キラ付け? Is it something about sparkling? (i've searched the wikia, yet I'm not still sure)

キラ付け means the task/work to get that effect, not just means the effect (キラキラ).
https://e-hentai.org/s/b3e69be10f/1111314-19
Mutsu is claiming to Teitoku that he made Ushio do the task without the supporting ships, and he's making an excuse for having accidentaly done that.

ファミスタもどき/famista-like (lesson)? I'm not sure it will fit with the story but I may figure it out if I see the source.

艶技
sex techniques? 艶's nuance depends on what they'll actually do in the lesson.
キラ付け means the task/work to get that effect, not just means the effect (キラキラ).
https://e-hentai.org/s/b3e69be10f/1111314-19
Mutsu is claiming to Teitoku that he made Ushio do the task without the supporting ships, and he's making an excuse for having accidentaly done that.

Its this one https://e-hentai.org/g/380652/f9414a98a1/

And do you know what's the ナデナデ stuff?

Its this one https://e-hentai.org/g/380652/f9414a98a1/

And do you know what's the ナデナデ stuff?

Famista, afaik, is short for "Family Stadium", a baseball video game series produced by Namco.
[en.wikipedia.org] https://en.wikipedia.org/wiki/Family_Stadium

Nadenade isn't just "撫でる"?

Its this one https://e-hentai.org/g/380652/f9414a98a1/

And do you know what's the ナデナデ stuff?

ナデナデする is a childish way of saying なでる (pat/stroke), and a phrase "ナデナデして くれたよ ω" is from the song.
https://www.youtube.com/watch?v=SiNezxfiiR8&t=118s
[www.project-imas.com] https://www.project-imas.com/wiki/Do-Dai

Namco.

That makes sense. However, unfortunately I'm not familiar with im@s so I can't fully understand the exact meaning of "famista-like".

BTW, the title is probably from [ja.wikipedia.org] https://ja.wikipedia.org/wiki/%E3%83%84%E3%...%83%88%E3%83%B3

This post has been edited by nonotan: Feb 23 2018, 16:16

Thanks for the help guys.

Just these colored bits.

Just these colored bits.

Green: Here, tits.
Purple: Boing (or whatever SFX you imagine when tits are revealed)