Skip to content

GuideMake and fix PDFs

Searchable

Searchable reads scanned PDFs with the text recognition built into Windows and saves copies whose words can be searched, selected and copied.

All chapters

When to use it

  • Scanned letters or client documents that are only pictures of words
  • Documents from the other side produced as image-only PDFs
  • Before making bookmarks from tab or exhibit pages that are scans
  • Scans saved as pictures (TIFF, JPEG or PNG) that need to become searchable PDFs
The Searchable screen: five scanned PDFs, each with the pages it will read, ready to be made searchable.

Before you start

  • Windows’ own text recognition. English comes with Windows set up in English. For other languages, add the language with Optical character recognition in Windows Settings, Time & language.
  • Scanned PDFs that open without a password, or scans saved as pictures: TIFF (every page), JPEG or PNG.

Step by step

  1. Open Searchable

    Click Searchable under MAKE & FIX PDFS in the side bar.

  2. Click Add scans

    Or drop PDFs, pictures or a folder on the window. Each row soon says how many of its pages need reading, such as “12 pages to read”, or “Already searchable”.

  3. Choose the language

    Under LANGUAGE OF THE SCANS, choose English or another language Windows can read on this computer.

  4. Check SAVE INTO

    Copies go into a folder called Searchable copies beside the first file unless you click Change.

  5. Click Make searchable

    With several files it reads, for example, Make 4 searchable. The header shows which file and step it is on.

  6. Read each row

    Each row ends with a word such as Made searchable, Already searchable, Could not or Failed checks, with lines saying what was read and checked. Show opens the folder the copy is in. A results screen accounts for every file, with Open the folder, Open the PDF, Send to… and Back to the list.

Settings worth knowing

Settings pane3
Read every page again, even ones with text
Off to start with. For a PDF whose text is wrong or only partly there. The text a page has stays, and the words read are added as well, so a search may find some words twice.
LANGUAGE OF THE SCANS
English, or any other language Windows has text recognition for on this computer.
SAVE INTO
The folder for the searchable copies; click Change to choose another.
⋯ menu (SETTINGS)3
Make these my defaults
Keeps the language you chose for next time.
Back to how the program came
Puts the language back to English, with an Undo.
Clear the list
Takes every PDF off the list.

What you get

  • A copy of each scanned PDF or picture named “… (searchable).pdf” in the folder you chose (Searchable copies, beside the first file, unless you change it). A name already taken gets a number.
  • The pages look the same; the words read are on them as an invisible layer, so you can search, select and copy.
  • No copy for a PDF whose every page already has text: “Every page already has text, so nothing needed reading. No copy was made.”
  • A copy that fails its checks is named “… - FAILED CHECKS.pdf”.

What Sudex checks for you

  • Each page is looked at on its own. Pages that already have text are left exactly as they were, and checked to be unchanged; a scan whose only text is a stamp, such as a Bates number or CONFIDENTIAL, is still read, and the stamp stays as it was.
  • Every word read is checked to be in the copy, in order.
  • Each page that was read is drawn again and compared: it must look exactly as the original does.
  • The copy must have the same number of pages, and the original is read again afterwards to confirm it is unchanged.
  • Pages where nothing could be read, pages that look upside down or on their side, pages where much of what was read looks like stray marks, and scans that look like another language are pointed out.

Tips

  • Turn sideways or upside-down pages the right way up in Pages first, then make the file searchable.
  • Nothing leaves this computer: the reading is done by Windows on this machine.
  • If the language you need isn’t in the list, add it in Windows with Optical character recognition; it then appears here.
  • A copy whose name ends “FAILED CHECKS” should not be relied on.

If something goes wrong

It says nothing can be read on this computer

The settings pane says “This computer cannot run Windows’ text recognition, so nothing can be read here.” and rows show Not here. Searchable needs Windows’ own text recognition; if your Windows is set up in English and this still shows, use Copy details for support in Settings and email [email protected].

Accented letters came out wrong

The row says the scan “looks like …, but was read as English”. Add that language in Windows Settings with Optical character recognition, choose it under LANGUAGE OF THE SCANS, and make the file searchable again.

Some pages weren’t read

“Nothing could be read on page …: a photograph, blank, or too faint.” For pages that “look upside down or on their side”, turn them the right way up in Pages, then make the file searchable again.

A file wouldn’t go in

Only PDFs and scanned pictures can be added; others are named in a message such as “Not added: … (it is not a PDF or a scanned picture).” A PDF that needs a password is marked Needs a password: “It needs a password. Save an unprotected copy and use that.”

It says Already searchable, but the text is wrong

Turn on Read every page again, even ones with text, and make it searchable again.

Picture at full size