Text, or a picture of text
A PDF either stores real text, or it stores a picture of text. Export one from Word, Google Docs, or a billing system and the letters are characters you can highlight, search, and paste. A scan from a copier, or a photo dropped into a PDF, is the other kind. The page is an image. The computer only sees pixels.
You can tell them apart before you upload anything. Drag across a sentence. If each word highlights on its own, the file has a text layer. If the whole page selects as one block, it is a picture. Search for a word you can see. A jump to that word means text. No hit means a scan. Nothing you upload can invent characters the file never stored.
What real text can become
When the words are real, you can copy them, search them, and turn the file into a document. PDF en texte reads the stored text and returns a plain .txt file. Fonts, bold, and pictures stay behind. A table or a two-column page can come out as one stream, so check the order where the layout carried the meaning. A locked file has to go through Déverrouiller le PDF first, and only if you are allowed to remove the password.
PDF vers Word rebuilds that text as an editable DOCX, up to 50 MB. A plain letter is often ready to edit. A newsletter, a form drawn as lines, or type sitting on a graphic can shift. The PDF goes over an encrypted connection, is converted in a temporary folder, and is deleted when the DOCX comes back.
What a scan actually contains
A scanned PDF is a photograph inside a PDF wrapper, so copy and search have nothing to hold. PDF to Text stops, because it only reads a text layer. PDF to Word cannot rebuild paragraphs from a picture. Recognition has to look at the image itself.
Image en texte does that reading in the browser, and it accepts an image, not a PDF. PDF en JPG renders the pages on your device first. Standard is 110 DPI, High is 180 DPI, and Very high is 290 DPI, which keeps small type sharper. Choose the language of the page so accents survive. Clear, well-lit print usually comes back almost intact. Handwriting and a low-resolution photo are harder. The scan is left unchanged, so it is still not searchable.
Pages that only half copy
A mixed file feels broken because some pages copy and others do not. The body was exported from a document, then a signed page was scanned on, or a screenshot was pasted in. PDF to Text and PDF to Word can use only the real text. The picture contributes nothing until that page is rendered and read with Image to Text.
A scan is not a damaged copy of the digital file. Someone printed it and sent a photo back. The converter then has a picture, which is why the Word file comes back holding an image instead of paragraphs.
Why the way back is harder
Saving a document as a PDF keeps the text. Turning that PDF back into a document has to guess the layout. Word vers PDF starts from real characters and keeps the layout, fonts, images, and tables, up to 50 MB. An unusual font may be replaced by a metric-compatible one. The new PDF can be searched because the words were characters.
A PDF records where ink was placed. It does not have to record that a block was a heading, a table, or two columns. PDF to Word rebuilds those pieces as well as the file allows. A designed page may reflow. Converting the same file again adds nothing the PDF never contained.
Searchable means a new file
Getting the words out is not the same as making the original PDF searchable. Image to Text leaves the scan untouched. Paste the checked text into a document and export that if someone needs a file they can search. A PDF saved from real text has a text layer. Check every name, date, and decimal against the paper before anyone else relies on it.
Size, locks, and where the file goes
A scan is often large because every page is stored as a picture. Compresser un PDF can shrink it enough for an upload form, and the smaller file is still a picture. Image to Text, PDF to Text, and PDF to JPG stay in the browser. PDF to Word and Word to PDF are sent encrypted, converted in a temporary folder, and deleted when the result comes back. PDF to Text and PDF to JPG refuse a locked file until you unlock it.
Questions fréquentes
Why can I see the words and still not copy them?
The page is a picture. The characters were never stored, so there is nothing to select.
Will Image to Text make my PDF searchable?
No. It puts the words in a box and leaves the PDF as a picture. A searchable file has to be exported from real text.
Can PDF to Word convert a scanned contract?
No. A scan has no text layer. Read the page images with Image to Text, then check the figures.
Which tools upload the file?
Image to Text, PDF to Text, and PDF to JPG stay on your device. PDF to Word and Word to PDF are encrypted, converted, and then deleted.