Free online tool — no sign-up
Upload a Chinese EPUB and get the first fifteen pages back in Japanese. Chinese tells you nothing about how two characters stand to each other, and Japanese cannot finish a sentence without knowing.
The first 15 pages of text are sent to our server to be translated. The file itself stays on your device.
Every Japanese verb ending is a statement about the relationship between the speaker and the listener. Chinese supplies no such information — its pronouns and verbs are flat on this axis — so the translation has to reconstruct it from who the characters are, what they do and how the story treats them. From a single paragraph that is impossible; from a whole book it is straightforward.
The shared script is the other thing to be careful about. Chinese and Japanese use many of the same characters with different meanings, so a compound cannot simply be carried over even though it would look perfectly natural on the page. The translation goes through meaning, not through the characters.
The model reads the entire text first and writes a guide — the cast, their relationships, the names and the recurring terms — and each page of the sample is written against it.
A DRM-free EPUB. The file stays on your device; only the text of the sample is sent.
The cast and their relationships are mapped once, which fixes the speech levels for the whole sample.
About 4,500 words as a complete EPUB, with kanji, hiragana and katakana used as Japanese uses them.
Speech level is inferred. The Chinese never stated it. The guide makes the inference consistent; it cannot make it authoritative.
Shared characters are not carried straight across. A compound that exists in both languages often means different things, so the translation follows the meaning even when the characters would have looked plausible.
Fifteen pages, not the book — about 4,500 words — and the text is sent to a server and discarded when the job ends.
From the relationships mapped across the whole book. This is the clearest reason the model reads the entire text first: Chinese does not encode the relationship anywhere, so a per-paragraph translator re-guesses and the characters' footing changes every few pages.
With their Japanese on'yomi readings, which is the convention for Chinese names in Japanese text — and the same name gets the same reading everywhere, which is what the guide is for.
Yes, both simplified and traditional. Neither changes how the text is read.
About 4,500 words, roughly fifteen pages, five books a day.
All free AI book tools · Free converters that run in your browser
The app runs the same engine across the entire book and can show the Japanese next to the Chinese, with your position synced between devices.
Open DualBook