How to ACTUALLY turn off Auto OCR - or I truly despise 'performing page recognition' / Automatic OCR

[Clint__]

For myself (and it appears many others) this is the worst feature Adobe Acrobat (maybe Adobe period) currently has. Trying to edit a PDF is a nightmare with this idiotic scan. It repeatedly scans, over and over (once isn't enough?) it takes forever, and oddly it tilts random images off center.

What is the purpose of this? To frustrate users into using another PDF editor? I could understand if it was a work in progress, but you can't even turn the damn thing off. I've tried all the "steps" recommended and nothing. Yes, I have the latest version. 

Threads like this appear through-out Adobe support. It's currently a garbage feature - allow me to turn it off until it's useful. Until then, I'm using another editor. 

[Veronique33831108uh8i]

Thanks for the tip. I did this and also barb's suggestion and it stoppped finally!

[Nora_Nora]

OH wow I did not realize the page recognition is different from the text recognition. no wonder the checkbox is useless....

[Nora_Nora]

And thank you Kym5E5E, I've just tried your suggestion, fingers crossed! Do you happen to know what those settings do?  I don't really understand what the reading order even refers to.  

Thanks

[Kym5E5E]

I don't actually know what those settings do but I've never noticed a difference without them.

[BarbLovesInDesign]

I think - don't know for sure - it refers to screen readers being able to convert 'text' to speech. I guess that would involve text recognition before they could 'speak'.

[Cherelyn32892709zbbp]

Reading order is the direction it is going to recognise your text. From left-to-right top to bottom is the way you normally read an English language book. Some other languages go right to left.

[Someolderguy]

and when the program locks up, I get a survey to rate whether I will recommend Adobe.  

[winmark01]

Dear all,

First, to Bernd Alheit: your response does not fit the issue, and you need to look again.

First, in my version of Acrobat Pro DC (Mac 2015.006.30527), there is no 'Recognise text' button to turn on or off in the Edit PDF side-bar menu, regardless of what you and the user guide claim. (see my two attached images)

Second, having text recognition and saving occur automatically is a very sub-standard feature – I am a happy advocate for Adobe products usually, but Adobe have screwed up on this one. It needs fixing.

Third, the situation that brings me to this forum is that I recently exprienced the following; I often check and pre-flight files for a publisher - files supplied by other designers – before they go off to the printers. In the last fews days a disappointed author has come back with complaint to the publisher that a small part of one image, which contained a scan of some [intentionally] hand-drawn lettering got printed as ugly gobble-de-guk. This seems to have occured through automatic 'text recognition' process forced on all pages when I made one intervention on another image on another page.

Now the question of liability raises its ugly head! No-one is happy.

Bernd Alheit – you may think you have correctly answered the issue, but you haven't really – please look again and show you expert credemtials by coming up with solutions that work!

I'm with the other complainats on this one. Simply not good enough from Adobe.

Cheers

Bruce

[Bernd Alheit]

The option is available in Adobe Acrobat DC:

[winmark01]

Hi Bernd Alheit, thanks for responding, but reiterating a ’solution’ that doesn’t work simply makes things worse, particularly as your advice is marked as a ‘correct answer’.

As I mentioned in my earlier post, and sent you screen grabs of, the 'Recognise text' button DOES NOT APPEAR at any time in my version of Acrobat Pro DC (which I am informed is up-to-date) in the 'Edit

PDF' panel. The rest looks the same, but the command you keep referring to is simply not there.

In the place where you show a tickable 'Recognise text' button, I simply have a curved arrow with a 'Revert to image' label – this is greyed out and only active when an image has been converted.

Please tell me, if you know, what is going on, and why the discrepancy?

[The version I have is 2015.006 30527] running on Mojave.]

[James307138546u2t]

I have same issue with "solution" not working, too. Painful!

[BarbLovesInDesign]

this feature is under the EDIT menu, not as edit under Print Production/Edit.

[Scott31841604pees]

I couldn't agree more. Whoever thought having this automatically happen is a good thing didn't use common sense.

[Abuelo_Duane]

I work with a lot of hand-written documents and this is a royal pain. I want it to stop on all pdf's, not just each one I am working on. As soon as you click the Edit PDF to turn off recognize text, recognize text has already started to mess up the text.

[Abuelo_Duane]

It's even happening when the recognize text button is unchecked. Where is the permanent solution?

[Marilyn327158208g1a]

OMGOSH yes! This is the most idiotic feature. It is making a challenging task all the more challenging by tilting pages with images. I can't get it to stop and sometimes can't get the images straight. What is the point of this feature? I was hoping to get an answer for how to turn it off. I hope someone from Acrobat reads these. But they already have our money so probably not. 

[QINGCHARLES]

It has been two years and they still haven't fixed it, so I guess they're happy with it.

[lindav11859390]

Sorry to reanimate a 3 year old post, but this is the only one I've found which really lays into this issue as it should.

As someone who mostly uses Acrobat to compile things like artist residency applications, this is actually so infuriating. The documents I try to compile in Acrobat consist of a mix of a few other pdfs, namely:

1. Image compilations I've made in Bridge (so combining my artwork psds into a pdf)

2. My CV

3. Motivation letters, answers to questions from the open call, etc.

The OCR somehow thinks that literally little bits of linework and texture in my illustrations, which bear no resemblance to letters, are letters and it tries to convert them. I've tried every solution presented in this discussion and still nothing works. And I cannot uncheck "recognise text" under scanned documents because my documents are not scanned! They're all created digitally.

Genuinely Adobe, please make this less painful for us. We already pay you way too much.

[pavelb72342004]

Yeah, it is the PAIN in all my body parts, when Acrobat tries to recognize 200+ pages of the document because I accidentally clicked the edit button. 
Adobe, please disable this useless auto "feature", your users are smart enough to click the "recognize" button in case we need it.

[Eric5FDD]

NOTE TO ADOBE: Not allowing users to turn OFF default automatic OCR when trying to edit a document is not enhancement of users' productivity - which is what your users pay you for - but theft of productivity. I must constantly click the "undo" error or "esc" key each time I wish to edit a scanned pdf, when sometimes I wish only to add a single line of text on top of an image.

You are aware of this conceptual and programming defect, but have not corrected it to enable users to disable auto OCR. Perhaps you believe auto OCR gets it right all the time, but it does not. Numerous errors result when it attempts to decipher moderate to low quality scanned documents, especially when there are images or signatures. This is understandable, but then please allow users to disable the auto OCR feature at their discretion. Thank you.

[Dustin34194422zzp1]

go to preferences, select "reading", at the top of that page select "Do not read documents"    Hope that helps.  SUPER ANNOYING, I know...

[Sahajah]

Just another unwanted and useless function forced on us by Adobe.  Stop pretending you are clever and leave my documents and my program the way I have it configured.  Your arrogance and intrusion into my decisions is intolerable and purely fascist.

[Dan5E2E]

Compounding the issue is the newest version of Acrobat DC and the rearranged "even smarter" interface decisions and new locations to learn. Here's how I managed to unselect and remove text recognition, starting with the new location of the OCR default checkbox in verion 2024.001.20643.

 I loaded a PDF of a line drawing with three hand-written words on it. 

Clicked Edit. The screen went black.

Clicked the gear/settings button, unchecked the Recognize scanned text.

An alert box opened saying that I'd lose any unsaved changes. EXACTLY. Now they get it!

The image reappeared as originally loaded. No OCR.

NOTE: My edit was simply to crop the image. No desire to add text.

Before doing this, I also tried the Reading preference settings suggested above. That may have prevented automatic scanning when I opened it.