Connecting to Sapphire
Aligning the stars
Open your account in minutes and start trading with the tools, insights, and support you need to make informed decisions.
Extract printed text from an image using OCR that runs on your device.
Drop your files here
or browse, or paste from clipboard
JPG, PNG, or WebP images
Processed in your browser - never uploaded
Image to Text (OCR) pulls the printed or typed words out of a picture and hands them back to you as text you can actually edit. Point it at a photo, a screenshot, or a scan in JPG, PNG, or WebP format, and it reads the characters using OCR (optical character recognition) so you don't have to retype anything by hand.
The entire process runs inside your own browser, on your own device. Your image is never uploaded to a server and never leaves your computer or phone, which makes this a safe way to lift text from documents that contain personal or sensitive details. The one thing to expect: the very first time you use the tool, your browser downloads a small OCR engine (about 5MB). After that it is cached, so every run afterwards starts instantly.
OCR works best on clear, high-contrast printed text, and it will happily read a clean screenshot or a sharp scan. Be realistic about its limits, though: handwriting is unreliable, and blurry, dim, skewed, or low-resolution images can produce garbled results. When that happens, a better source image almost always fixes it, and you can tidy up any stray characters in the editable output before you copy or save.
Open the Image to Text tool
Open the tool in your browser on any device. The first time you run it, your browser quietly downloads a roughly 5MB OCR engine and caches it, so later visits start straight away.
Choose your image
Select or drag in a JPG, PNG, or WebP file - a photo, a screenshot, or a scan. Everything stays on your device; the image is never uploaded anywhere.
Let the OCR read it
The tool scans the image on your device and detects the printed or typed characters. Clear, high-contrast text is read quickly and accurately; blurry or low-quality images take longer and are less reliable.
Review and edit the text
Read through the extracted text and fix any characters the OCR misread, especially around faint, skewed, or handwritten areas. The output is fully editable, so corrections take seconds.
Copy or download the result
Copy the text to your clipboard for pasting elsewhere, or download it as a .txt file to keep. No account, watermark, or sign-up required.
The OCR runs entirely in your browser using a self-hosted engine, so your images and the text inside them are never uploaded to any server. That makes it genuinely safe for screenshots of receipts, IDs, statements, or anything private.
You get plain, editable text you can copy or download as a .txt file, with no cost, no account, and no watermark. Retyping by hand is gone - and you can correct any OCR slip-ups right in the box.
The one-time ~5MB engine download is cached by your browser, so every extraction after the first one starts instantly. Once it's cached, the tool keeps working even without a connection.
Feed it JPG, PNG, or WebP files straight from your camera, screen captures, or a scanner. There's no need to convert anything first.
Most online PDF tools upload your document to a server to process it. This one does not. Every operation runs inside your own browser using WebAssembly, so your files are read, processed, and saved locally - they are never transmitted, stored, or seen by Sapphire Broking or anyone else. That makes it safe to use with financial statements, contracts, and identity documents.
Keep going
Shrink JPG, PNG, and WebP images to a target size or quality.
Open toolChange an image's dimensions by pixels or percentage.
Open toolCut an image down to the region you want with a drag-to-select box.
Open toolRotate and flip images by 90-degree turns.
Open tool