Skip to main content

OCR Text Recognition

Recognize English text (letters, digits, symbols) from scanned PDFs and images using AI (DBNet+CRNN). Japanese, Chinese, and other scripts are not supported. All processing happens in your browser — no data is sent to any server.

Drop an image or PDF, or click to select

Supported formats: image/*,application/pdf,.pdf · Max size: 100MB

Japanese text recognition is not supported (English only)In-browser AI cannot yet recognize Japanese accurately enough to meet our quality bar, so we keep this tool English-only rather than return unreliable results. To pull out Japanese text already embedded in a PDF, use the Text Extraction tool instead.Open Text Extraction →

Chrome / Edge is recommended.

How to use

  1. Drag & drop a scanned PDF or image

  2. AI detects text regions and recognizes the text (English only)

  3. Copy the result or save it as text

Frequently asked questions

Can it recognize Japanese?
No — English only. In-browser AI models cannot yet recognize Japanese accurately enough to meet our quality bar, so rather than return unreliable results we keep this tool English-only. If the Japanese text is already embedded in the PDF, use the PDF Text Extraction tool. For scanned Japanese documents, consider on-device OS features such as iOS Live Text.
Which browsers are supported?
Because the AI runs in your browser, Chrome or Edge is recommended. Images are never uploaded.