Exporting to Word shows junk characters

Forum|Forum|8 years ago
April 14, 2017
3 replies
2847 views

I have a document in PDF, I want to convert into word. But Acrobat XI gives me some junk readable characters after converting. Can anybody please help. One of the file has given through Google drive.

Thanks in Advanced

Convert.pdf - Google Drive

Edit and convert PDFs

Karl Heinz Kremer

Community Expert

I replied by email, so did not see that Meenakshi already did add a reply regarding the nature of your document. Let me add a few more thoughts: Not only is this a scanned document, but it's also a scan of a document using a script font. These are exceptionally hard to recognize. It's also not the best scan quality:

As you can see, the text is pixelated, and has "smudges" which all make the OCR job a lot less reliable. I even ran your file through a dedicated OCR application (Abbyy FineReader), and not even that was able to produce any useable output. FineReader is my go-to application if I get to the limits of what Acrobat's OCR can do.

Chances are that - unless you can find an OCR application that supports the specific font that is used in your document - you will not be able to export this document to Word.

M

mike_nomad

Participant

Even after all these years of using Adobe Acrobat Pro, it is still the same (not helpful in OCR). I found other online sites to be able to do OCR from crappiest PDFs with great ease up to 90% accuracy, only you have to edit it.

AnandSri

Community Manager

Hi,

I hope you are doing well, and we apologize for the delayed response and the trouble.

If the PDF is scanned in high quality and the fonts were properly embedded, Acrobat should be able to recognize the text with ease. Could you share the file with us for further investigation? We're here to assist you.

You can also see this article for more information on the OCR: https://adobe.ly/4gQWuDc, and https://adobe.ly/4gNmX4T

Thanks,

Anand Sri.

Karl Heinz Kremer

Community Expert

Converting from PDF to Word, Excel or any other format is one of the most complex things you can try to do with a PDF file. It works very well in some cases, in other cases the output has very little to do with the original file. The key for success is that the PDF file needs to be "tagged" - which means that it contains information about the information that is displayed in the file. The best way to make sure that a PDF file is tagged correctly is by using the PDFMaker in Acrobat to create the PDF file from Word or Excel (that's the Acrobat ribbon or toolbar).

Unfortunately there is not much you can do to improve the output without spending a lot of time (e.g. by manually tagging the file). Also, if you are using Adobe's ExportPDF service and don't have access to Acrobat, that is not even an option.

The only thing you can do is complain to the original author of the file and tell them that they used a bad PDF generator to create the PDF file.

Sometimes it helps to save the PDF file as a set of high resolution (e.g. 600dpi) images, then import these images back into Acrobat, run OCR and then export to Word or Excel again.