How to OCR every page of a multi-page TIFF

Every page of a multi-page TIFF is read when you open it in Read by Zaweba, not just the first one. The pages appear in a list with thumbnails, the text for each is shown beside its picture, and you can save the lot as one text file, one .docx document or one searchable PDF.

Multi-page TIFF is an old workhorse. Fax software, office scanners and document archives have saved bundles of pages that way for decades, and many simple viewers and converters only ever show page one. If you've opened a TIFF and wondered where the other nine pages went, that's why.

Before you start

Read by Zaweba on a Windows PC and the TIFF file. TIFF is one of the formats Windows decodes on its own, so there's nothing extra to install. Read reads it entirely offline, which matters for the kind of material that tends to live in TIFFs: contracts, medical letters, old office records.

From file to text

Drop the TIFF onto Read's window, or open it from the folder it lives in. Read pulls out each page in turn and runs its clean-up on every one: turning sideways pages upright, straightening crooked ones, evening out shadows and lifting faint print where needed. Each step only acts when the page needs it.

For a multi-page document, Read's Auto setting uses the Tesseract engine, which does layout analysis and gives a confidence for every word. That means the review is properly useful. Pages are listed down the side with thumbnails; pick one to see its cleaned-up picture next to its text. Words under 70 percent confidence are shaded amber in both, and F8 jumps to the next one, across pages, so you can work through the whole document checking only the words that need it.

When you're happy, choose how to save. A .docx keeps page breaks, paragraphs, tables and two-column sections, with no need for Word. A searchable PDF keeps each page looking just as it did, with every word laid over it as invisible text, so Ctrl+F finds words and copying works. Your corrections in the review go into the text layer too. Plain text gives you the words alone.

Choosing a layout mode

Old faxes and forms often have columns, addresses and short lines. In Paragraphs and columns mode, Read joins wrapped lines into paragraphs, rejoins words split by a hyphen at the end of a line, and reads columns one after another, while keeping the line breaks in blocks of short lines like an address. If you'd rather keep every line exactly as it is on the page, switch to Single column, line by line.

Lots of TIFFs at once

For a folder full of them, use the Batch tab. Read works through every file in the background, in the formats you've ticked on the Output tab, and puts the results in a Read by Zaweba folder beside the pictures or in a folder you choose. You can have one combined document or one set of files per picture. Jobs queue up and show their progress and how many words need checking, and a file that can't be read is reported without stopping the others. Pages are kept as JPEG plus words while it runs, so memory stays low even on long jobs.

Expectations

Fax quality varies. Very low resolution pages, smudged print and handwriting are among the limits Read lists honestly in its Help tab, so expect more amber words on a poor fax than on a clean office scan. The review is there for exactly that, and Read by Zaweba gives you every page to check rather than just the first.