PDF to Text

Turn scanned or digital PDFs into plain text.

No languages match your search.

📄

Drag and drop, or paste a screenshot

Image, PDF, Word (DOCX) or text - drop, browse, paste, or use a URL

🔒 Your files are processed privately and deleted automatically.

OCR.chat extracts text from any PDF, whether it is a digital export or a stack of scanned pages. Upload a multi-page PDF and every page is processed and combined into one clean, copy-ready result. Pages that already contain a real text layer are read instantly and losslessly, while scanned pages are run through OCR automatically, so you always get the most accurate text the file can offer.

This PDF to text converter is made for people who work with long documents: researchers mining old papers, paralegals turning case files into searchable text, and analysts pulling figures out of scanned reports. It reads more than 100 languages, shows the extracted text beside the original page for review, and never silently translates foreign words, so a multilingual PDF comes out exactly as it went in.

There is no signup needed to try it, and the fast engine handles standard printed PDFs for free. When a document has dense tables, columns, or handwriting, the premium AI engine delivers higher accuracy. Download the result as plain text, Markdown, Word, CSV, or JSON, and rest assured your file is processed for OCR and then deleted automatically.

How to pdf to text

1
Upload your PDF
Drag in a single PDF or browse to it; multi-page documents are fully supported.
2
Choose the engine and language
Stick with the free fast engine for printed pages, or switch to the premium AI engine for scans, tables, and handwriting; set a language or auto-detect.
3
Let it sort the pages
Pages with a real text layer are read instantly and losslessly; scanned pages are OCR'd at high resolution, all merged into one result.
4
Review and export
Check the text against each page, then copy it or download as TXT, Markdown, DOCX, CSV, or JSON.

Common uses

  • Researchers extracting passages and citations from scanned journal articles and books.
  • Lawyers and paralegals converting scanned case files and contracts into searchable text.
  • Accountants pulling figures and line items out of scanned financial statements.
  • Archivists digitizing historical records and out-of-print documents for preservation.
  • Office workers reusing content from old PDFs without retyping page after page.
  • Journalists searching long government and court PDFs for specific names and terms.

Frequently asked questions

Yes. Every page is processed and the results are combined into one continuous text output you can copy or download.

Pages that contain a real text layer are extracted instantly and losslessly, with no OCR needed. Only scanned pages are run through OCR, which keeps quality high and speed fast.

You can try it for free, with per-page metering where credits equal pages. A free account adds a monthly page bucket, and paid plans from $5/month add more pages and batch processing for large documents.

Over 100, including CJK, Arabic, Cyrillic, and Indic scripts. Foreign words are transcribed as written and never auto-translated.

Yes. Scanned pages are rasterized and OCR'd automatically; the premium AI engine is recommended for faint scans, tables, or handwriting.

Your file is processed for OCR only and then deleted automatically. We never sell or share your documents.

Yes. POST your PDF to /api/v1/ocr/ and receive the combined text, optionally specifying language, engine tier, and output format.

Remove the password or unlock the PDF on your device first, then upload it. Once a file is open we can read it, but we cannot bypass encryption on a locked document.

The premium AI engine understands columns and reading order, so newspaper, journal, and two-column layouts come out in the right sequence instead of having the columns run together.

Batch processing is available on paid plans, so you can queue multiple PDFs in one go. On the free tier you upload one document at a time.

Plain text (TXT), Markdown, Word (DOCX), searchable PDF, CSV, and JSON. A searchable PDF keeps the original page images with a hidden, selectable text layer added on top.

No. Your upload is left untouched and the extracted text is delivered as a separate result, so the source file is never modified.

Use this via the API

Run this tool programmatically with a single POST. Authenticate with the API token from your account page.

curl -X POST https://ocr.chat/api/v1/ocr/ \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -F "file=@your-file.pdf" \
  -F "tool=pdf-to-text"

Files of 5 pages or fewer return the result inline; otherwise poll the job, then download it as txt:

curl -L "https://ocr.chat/api/v1/ocr/JOB_UUID/download/?format=txt" \
  -H "Authorization: Bearer YOUR_API_TOKEN" -o result.txt
Read the API docs →

More OCR tools

🖼️ Image to Text 📝 PDF to Word 🗎 Image to Word ⬇️ PDF to Markdown 📊 Table Extractor 🔍 Searchable PDF ✍️ Handwriting to Text ➗ Math to LaTeX 🧾 Receipt Scanner 💳 Invoice OCR 🧩 Document to JSON 🌍 Translate a Document 🔤 Word to Text 📃 Word to PDF 📈 PDF to Excel 🧮 Image to Excel 🌐 Image Translator 👁️ Describe Image ♿ Alt Text Generator 💬 Chat with PDF 🗨️ Chat with a Document
Rate this page
★ ★ ★ ★ ★
5.0/5 (0)

What could we improve? Your feedback helps us fix issues.