PDF ↔ TXT
Extract selectable PDF text or turn UTF-8 text into a PDF with automatic wrapping and page breaks. Korean, English, Japanese and Chinese are supported.
Your files stay on your device. All conversion happens in your browser.
This page is for text. Pull the text out of a PDF, or make a simple PDF from text you type or from a .txt file.
How to extract text, or make a PDF from text
- For PDF → text, add the PDF. Leave the Pages box empty for every page, or type 1-3,5.
- Press Convert. The text appears in a box you can copy, and you can download it as text.txt.
- For Text → PDF, switch the operation, then type or paste text, or add a .txt file saved as UTF-8.
- Choose A4 or Letter, a font size from 6 to 72 and a margin from 0 to 200 points. Convert and download text.pdf.
What to expect
- Extraction reads the text layer of the PDF. It does not read pictures of text, and it does not keep the original layout, so columns, tables and line breaks can differ.
- Text → PDF wraps long lines and adds page breaks by itself. Korean, Japanese, Chinese and English share one font that your browser loads from this site only when it is needed.
- A character the font does not contain cannot be written. The tool tells you to remove it instead of leaving a gap.
Measured example
We typed four lines in Korean, Japanese, Chinese and English, made a PDF (one A4 page) and extracted the text from it again: the result matched what we typed. A PDF that holds only a picture of a page answered “No text found. Scanned PDFs need OCR, which is not supported.” The PDF made from those four short lines was 7,693,860 bytes, because the whole font is stored in the file. Plan for that size if you want to send it by email.
Limits
- There is no OCR. A scanned page is a picture, so there is no text to extract.
- The layout is not kept. A table becomes lines of text.
- Text → PDF makes plain text pages: no images, no bold, no columns.
Questions about this page
Can it read a scanned PDF?
Not at present. OCR is not included, and a scan holds no text, only a picture of it.
Do Korean, Japanese and Chinese PDFs extract correctly?
Text that is stored as real text in the PDF comes out as it is. In our test, lines in all three languages and English came back unchanged. A PDF that draws its characters as shapes has no text to extract.