OCR — Image & PDF to Text

Beta

Drop images or PDFs here

PNG, JPG, WebP, BMP, TIFF, PDF up to 50 MB each

Files are processed locally — they never leave your browser.

Queue

No files yet. Drop images or PDFs above to get started.

Languages(1)English

Pre-processing

Grayscale

Auto-contrast

Binarize

Confidence filter≥ 60%

Words below this confidence are hidden from the output.

Output mode

Export

Add a file to get started

About OCR — Image & PDF to Text

OCR extracts text from photos, screenshots, scanned PDFs, and handwritten notes directly in your browser. Supports 100+ languages, auto-detects deskew and rotation, and can output plain text, Markdown, a structured JSON blob, or a searchable PDF with an invisible text layer. Table mode converts tabular regions into clean CSV you can paste into Excel.

How it Works

1Drop an image, PDF, or a batch of either.
2Pick the language(s) in the document.
3Optionally enable table mode or searchable-PDF output.
4Wait for the WASM OCR engine to process each file.
5Export as .txt, .md, .docx, .csv, or a searchable PDF.

Frequently Asked Questions

How accurate is it?

For printed text with decent contrast: 95-99%. Handwriting: 60-85%. Low-resolution phone photos: try the auto-deskew toggle for best results.

Does anything leave my browser?

No. The Tesseract WASM engine and all language data run locally.

Which languages are supported?

100+ including English, Spanish, French, German, Chinese (simplified + traditional), Japanese, Korean, Arabic, Hindi, Tamil, Russian, and more.

What is searchable PDF mode?

It outputs a PDF with an invisible, selectable text layer over the original image — perfect for archived scans.

About OCR — Image & PDF to Text

Command Palette

Queue

About OCR — Image & PDF to Text

How it Works

Frequently Asked Questions

Queue

About OCR — Image & PDF to Text

How it Works

Frequently Asked Questions