Skip to content

Copy text from an image

A free image to text converter that runs inside your browser. Drop a photo, scan or picture and convert the image to text you can actually paste. Nothing is uploaded, and there is no sign-up or watermark.

Drop a picture to copy its text

Photos, scans, menus, whiteboards, slides as PNG. Copy the whole image or select one region — it stays in this tab.

…or press Ctrl+V / +V to paste

100% local — files never leave your browser

How to copy text from an image

1

Drop the picture

A photo of a page, a menu, a whiteboard, a label, a slide exported as PNG. Drag it in, paste it, or pick a file — JPG, PNG, WebP, GIF, BMP and AVIF all work.

2

Convert the image to text — all of it, or just a box

The whole image is read locally by the OCR model. Need one paragraph? Drag a box on the preview, or click a block, and extract text from only that region.

3

Copy usable text

Plain text, a rebuilt table (Excel / Markdown / CSV), or a snippet you can paste. Optional cleanup joins broken lines without rewriting your words.

Why you cannot just select the text

It is pixels, not text

A JPEG of a contract or a PNG of a poster has no text layer. Clicking and dragging selects nothing. Converting an image to text takes optical character recognition — that is what this page runs, locally, to turn those pixels back into copyable characters.

Convert a region, not the whole photo

Menus, whiteboards and book photos are messy. Select the line you need instead of cleaning a wall of extra text. Most image to text converters give you the whole picture and leave the trimming to you.

OCR and layout models, running on your device

Recognition uses PaddleOCR PP-OCRv6 compiled to WebAssembly — the same class of model a server-side image to text AI would run, except it executes in your tab. Printed English, German, French, Spanish, Japanese, Korean and Chinese are all supported.

Tables come out as tables

Grid-like regions are rebuilt so you can paste into a spreadsheet, Markdown or CSV — not a jumble of spaces. Useful when the picture to text you need is really a price list or a results table.

The picture never leaves this tab

OCR runs in your browser. Your file and the extracted text are not uploaded or logged. Airplane mode works after the first load.

Frequently asked questions

Why can’t I copy text from a picture?

Images store pixels, not characters. There is nothing for the browser to select. This tool reads the picture locally and gives you real text you can copy.

Is this image to text converter free?

Yes, completely — no account, no page limit, no watermark and no paid tier. Conversion happens on your own device, so serving it costs us almost nothing and there is nothing to meter.

Does it use OCR or AI to convert an image to text?

Both descriptions fit. It runs PaddleOCR PP-OCRv6, a neural detection and recognition model, through ONNX Runtime in WebAssembly. The difference from a cloud image to text AI is where it runs: your browser, not our server.

Can I convert a picture to text online without uploading it?

That is exactly what this page does. The picture is decoded and recognised inside the tab. You can disconnect from the network after the page loads and the conversion still works.

Can I extract text from only part of an image?

Yes. After the preview appears, drag a box around the area — or click a block to auto-detect it — and convert just that region.

Can I copy a table from an image into Excel?

If the region looks like a grid, it is reconstructed as a table you can copy as Excel, Markdown or CSV. Proofread the cells; the plain-text view stays editable.

Which image formats are supported?

PNG, JPG/JPEG, WebP, GIF, BMP and AVIF. Screenshots pasted straight from the clipboard work too, and PDFs have their own page.

Which pictures give the best results?

Sharp, well-lit photos of printed text, flatbed scans and screenshots of slides. Handwriting, heavy blur, strong glare and very small type are noticeably less reliable in any OCR engine.

Need to copy from a screenshot or a PDF?