Typed text is the easy case for OCR. Handwriting is a genuinely harder problem -- and knowing why makes it much easier to get a good result out of a photo of your own notes.
Printed text is consistent: every 'a' in a given font looks like every other 'a'. OCR engines lean heavily on that consistency. Handwriting doesn't have it -- letter shapes vary from person to person and even line to line, cursive script often connects letters together with no clean break between them, and slant, pressure, and spacing all add variation a printed font never has.
That's true for every OCR engine, not just Scanquil's -- it's a genuinely harder recognition problem, not a solvable bug.
A few things make a real difference to the result: even, direct lighting instead of a shadow across the page; a flat, straight-on angle rather than shooting at a tilt; and dark ink on plain, light paper rather than faint pencil or a busy/lined background. None of this is Scanquil-specific -- it helps any OCR engine, because it makes the actual letter shapes clearer to read.
Choose your photo in Image to Text, and Scanquil reads it entirely on-device -- no upload. Download the result as a plain .txt file, or copy it straight out.
Messy handwriting or a low-quality photo (poor lighting, skewed angle, blur) is exactly where the free on-device engine is most likely to misread words. Image to Text (AI) reads the same photo with a different engine that generally handles these harder cases noticeably better.
The free Image to Text tool reads your photo entirely in your browser, on your own device -- it never leaves your computer.
It varies with how the handwriting is actually formed -- clearer, more separated lettering reads more reliably than fast, heavily-joined cursive. This is a hard case for any OCR engine, and the AI-assisted tool generally does better on it than the free pass.
JPG, PNG, WebP, and GIF.
Yes -- it works in any modern browser on a phone, tablet, or desktop, with no app to install.