InfyTool
Free icons

Image to text (OCR)

Photos of documents, screenshots, whiteboards — read into editable text by an OCR engine running on your own device.

Photos of documents, screenshots, scans · text is read on your device, nothing is uploaded

No size limit · files are processed on your device and never uploaded · how it works

Auto-detect is on — just drop an image and it reads English + Gujarati + Hindi together. First use downloads the engine once (≈6 MB + each language, 1–2 MB); after that it works offline.

Advanced settings (optional)
Recognition
Languages

Auto reads Latin + Gujarati + Devanagari together — correct for most documents with no picking. For other languages or a faster read, switch off Auto and choose exactly.

Image enhancement

Grayscale, contrast-stretch and upscale before reading — big accuracy gain on phone photos. Turn on Threshold for faint or low-contrast text.

Sections

Group the text into sections (paragraphs) with a confidence score, and hide the low-confidence noise.

Runs in your browser. Files are never uploaded — you can go offline after the page loads.

How to use the Image to Text (OCR)

  1. Drop photos, screenshots or scans — several at once is fine.
  2. Pick the language; the OCR engine reads each image on your device, with a confidence score.
  3. Copy the text or download everything as one .txt.

Easy to use

Drop files, pick options, download. Whole folders at once, no signup, no daily limit.

Best quality

The same engines the professionals use — SVGO, resvg, MozJPEG, libwebp, libavif — compiled to run in your browser.

Private & secure

Files are processed on your device and never uploaded. Close the tab and nothing remains anywhere.

OCR is exactly what shouldn't be uploaded

Think about what people OCR: identity documents, bills, contracts, prescriptions, salary slips, whiteboards from internal meetings. Every other OCR site ships those images to a server. Here the Tesseract engine — the most-used open-source OCR in the world — runs as WebAssembly inside your browser tab. Load the page once and you can switch off the internet and keep extracting.

Getting the best read

OCR loves flat, sharp, well-lit printed text. Photograph the page straight-on and fill the frame; screenshots read near-perfectly as they are. Each result carries a confidence score, so you know when to trust it and when to re-shoot. Handwriting is honestly out of scope — that needs cloud-scale models. For scanned PDFs, use PDF to Text, which runs this same OCR across every page at 300 DPI; for text living in a Word file, the Word converter extracts it without OCR at all.

Frequently asked questions

How accurate is it?

Honest answer: very good on clear printed text — documents, books, screenshots, signs — and each result shows its confidence score. Handwriting is not supported (that genuinely needs cloud-scale models), and low-light phone photos read worse than flat, well-lit ones. Straighten the page and fill the frame for best results.

Which languages can it read?

English, Hindi, Spanish, French and German — each a compact on-device model (1–3 MB) that downloads once when selected and then works offline. Tell us which language you need next.

Are my documents uploaded for OCR?

No — and for OCR that matters more than for any other tool, because the things people OCR are IDs, contracts, bills and medical papers. The Tesseract engine runs entirely in your browser; you can go offline and keep extracting.

Can it read text from a scanned PDF?

Yes — use PDF to Text: when it detects a scan, it renders each page at 300 DPI and runs the same on-device OCR across the whole document.

PDF to TextEvery word out of a PDF and into your clipboard or a Word to HTML & TextA Image ConverterPNG, JPG, WebP, AVIF — any direction

Your files never leave your device

Most online converters upload your file to a server. InfyTool ships the actual encoders to your browser as WebAssembly instead, so processing happens inside your tab — faster for batches, and nothing to delete afterwards.

  • No upload, ever. Turn off Wi-Fi after the page loads and every tool keeps working.
  • No limits. No file-size cap, no queue, no watermark, no account.
  • Runs on your CPU. Batches use several cores in parallel; nothing is throttled.

More free tools

Need graphics to work with?

InfyTool is built by the same team as these free libraries.

InfyPNGFree transparent PNG graphics for mockups and social posts. InfySVGFree SVG designs and illustrations, ready to optimise or convert. InfyIconFree SVG and PNG icons you can open directly in these tools.