Skip to content
Use case

Converting a Japanese PDF to EPUB

Updated:

Japanese text has no spaces between words, so joining lines and paragraphs needs different rules than Latin text. On a Japanese Wikipedia article the EpubCC engine joins CJK lines without inserting spaces.

Try 5 pages free

Original and result

CJK text without spaces between words, mixed with Latin script and numbers.

Original PDF
Page of the original PDF
Result (EPUB, as a reader shows it)
The same passage in the converted EPUB

Download source PDF Download the sample EPUB

Source and licence

Document
Wikipedia (Japanese): The Tale of Genji
Licence
CC BY-SA 4.0 (text); images under their own licences
Pages
67
Traits
text PDF, one column, Japanese, CJK

Wikipedia contributors, “源氏物語” (The Tale of Genji), Japanese Wikipedia, text licensed CC BY-SA 4.0; images under their own licences.

Settings used

  • Engine: EpubCC engine, build of 2026-11-03 (the one measured in the benchmark).
  • Pages: the whole document.
  • Options: none. The engine has no per-document settings yet; the same defaults apply to every file.

Validation and results

EPUBCheck EPUBCheck on our EPUB: 0 errors, 0 warnings.

8-gram recall · Like 4-gram recall with 8 words: drops when reading order or line joins break sentences (continuity). Methodology
Engine8-gram recall
EpubCC engine (EPUB output)0.963
EpubCC engine (plain text)0.963
pdf_oxide · extract_text0.970
pdf_oxide · to_plain_text (XY-cut)0.970
pdf_oxide · Markdown0.797
PDFium · text().all()0.952
OpenDataLoader PDF (local mode)0.942
pdftotext (Poppler)0.965

Limitations we observed

  • The source is horizontal text; vertical (tategaki) books have not been tested yet.
  • pdf_oxide and pdftotext score slightly higher than our engine on this document.

Convert a similar PDF

Share