Highlight PDF text and get summary of it in selectable text form

[letsgoslowbro 0]

Would be amazing to highlight PDFs (selectable text) within Evernote and have that highlighted text be aggregated into a summary in the note the PDF is in for example.

I currently use Good Reader app for this, but if Evernote made this obsolete I'd be very happy.

[PinkElephant 8,750]

Currently the EN OCR will not build a text layer that can be extracted.

The OCR does build a table of words, and where in the document they can be found. You can visualize this by a list of words, followed by the coordinates where this word is found. This process is server based, not running within of the clients.

Personally I don't expect EN to add a full OCR in the closer future. This would mean as well that EN would modify the documents the users store in their notes. GoodReader does alter the document, and it needs to be saved again.

You can send your input to EN PM via the feedback option, or contact support through a ticket.

[romelden 30]

在 2022/3/24 在 AM1點39分, PinkElephant說:

Currently the EN OCR will not build a text layer that can be extracted.

The OCR does build a table of words, and where in the document they can be found. You can visualize this by a list of words, followed by the coordinates where this word is found. This process is server based, not running within of the clients.

Personally I don't expect EN to add a full OCR in the closer future. This would mean as well that EN would modify the documents the users store in their notes. GoodReader does alter the document, and it needs to be saved again.

You can send your input to EN PM via the feedback option, or contact support through a ticket.

Agreed on this.  Actually EN can open the PDF directly by the others software. (Not always workable in EN 10).  It can be simple covert the PDF in others software and save it back very easy.  The words recognition in EN searching is powerful which instead of selectable text.  OCR do not fully 100% accurate, and also very limitation on multi-languages. So this must be very careful if it can modifying the documents in EN. 

[PinkElephant 8,750]

The easiest way is to create the pdf with an extractable text layer. Many PDFs already hold it, for example those created from Office documents. For others using a scanner with OCR software can be a solution. And finally there are apps that provide a text readout later.

Personally I try to OCR everything before uploading. Just keep in mind that EN will not make a second OCR when there is already an embedded text layer. So OCR quality should be good when using this strategy, otherwise search quality will suffer.