How to Turn a Photo of a Document Into Editable Word Text
You have a photo of a printed letter that needs a few changes, a screenshot of a policy you want to quote, or a scanned page saved as a JPG. A picture of text cannot be edited, but text recognition can turn it back into words you can work with. There are two routes, depending on whether you need just the words or a whole Word document.
Quick route: copy the text
Open Image to Text and add the picture from your device, Google Drive, Dropbox or OneDrive. Receipts, screenshots and photos of pages all work, in JPG, PNG, HEIC and other common formats. Choose the language in the picture and press Get the text. The words appear on screen, ready to copy into Word, an email or a note.
This is the fastest way when you need the words and not the layout, for example to quote a paragraph or fix a few lines and retype the rest yourself.
Full route: a Word document
When you want a .docx file with the text in paragraphs, go through a PDF. First, turn the photos into one PDF with Image to PDF, one page per photo. Next, run OCR PDF on that file, choosing the document's language, so its pages get real text. Finally, convert the searchable PDF with PDF to Word and open the .docx in Word or Google Docs.
Simple layouts, such as letters and plain reports, come through well. Complex layouts with columns, tables and text over images need more tidying in Word afterwards.
Better photos, better text
Text recognition is only as good as the picture. Photograph the page straight on, in good light, filling the frame, without shadows across the text. A dim or blurry photo can often be rescued with Image Enhancer before you run text recognition. Handwriting is read much less reliably than print.
Check before you use it
Read the result against the original, especially numbers, names and dates. Text recognition can mix up similar characters, such as 0 and O or 1 and l, and a wrong digit in an amount or a reference number matters far more than a typo in a sentence.