Blind and Visually Impaired Community
↑
2
↓
How do you use OCR on NVDA? (self.Blind)
submitted by ChipsAhoiMcCoy
means?Doesthismeanthatitsimplydidn'twork?THanks!
Emmenias 2 points 1y ago
Here's spaces added, for everyone's convenience:
"means? Does this mean that it simply didn't work? Thanks!"
Still looks like the main part of the post has been cut off. OP, could you elaborate (with spaces, preferably)?
But in general, make sure to maximise the window before using OCR, such as with windows +up arrow. If trying to OCR an image on a webpage, first bring your mouse's focus to it (NVDA +numpad divide on the desktop keyboard, and I think NVDA+shift+m on laptop). Remember that you cannot use OCR while screen curtain is enabled.
"means? Does this mean that it simply didn't work? Thanks!"
Still looks like the main part of the post has been cut off. OP, could you elaborate (with spaces, preferably)?
But in general, make sure to maximise the window before using OCR, such as with windows +up arrow. If trying to OCR an image on a webpage, first bring your mouse's focus to it (NVDA +numpad divide on the desktop keyboard, and I think NVDA+shift+m on laptop). Remember that you cannot use OCR while screen curtain is enabled.
TechnicalPragmatist 2 points 1y ago
Hahahaha! Thanks for the version with spaces hahahah!
GreatSunJester 1 points 1y ago
Press NVDA + r to trigger OCR. All you will hear is "Recognizing. Result document". There is now a virtual and invisible overlay on the scanned window. Very few NVDA keystrokes will function.
Press Escape to exit this mode.
You can use the basic reading keys (up and down arrow and the number pad for characters, words and lines).
You can also press Control + A to select all recognized text, Control + C to copy it and then paste it into another document.
Whatever was recognized in the image or PDF will be read, if anything. Nothing outside the window like title bars, buttons, or any part of the document which is not in view.
Best results seem to be in Acrobat. I opened a PDF of a magazine cover in both Edge and Acrobat. I did not allow Acrobat to do it's own OCR and made sure there was no real text present. With both PDFs at the same zoom level and visually the same size on my screen, NVDA was able to read more from the Acrobat window than Edge.
Do not expect miracles. As a test, I did the same test with JAWS. It's OCR was superior (as expected since JAWS basically has OpenBook integrated into it now). However, the resulting text was pretty useless.
Press Escape to exit this mode.
You can use the basic reading keys (up and down arrow and the number pad for characters, words and lines).
You can also press Control + A to select all recognized text, Control + C to copy it and then paste it into another document.
Whatever was recognized in the image or PDF will be read, if anything. Nothing outside the window like title bars, buttons, or any part of the document which is not in view.
Best results seem to be in Acrobat. I opened a PDF of a magazine cover in both Edge and Acrobat. I did not allow Acrobat to do it's own OCR and made sure there was no real text present. With both PDFs at the same zoom level and visually the same size on my screen, NVDA was able to read more from the Acrobat window than Edge.
Do not expect miracles. As a test, I did the same test with JAWS. It's OCR was superior (as expected since JAWS basically has OpenBook integrated into it now). However, the resulting text was pretty useless.
ChipsAhoiMcCoy [OP] 1 points 1y ago
I haven’t used draws quite yet, but I do have a license laying around that I need to redeem after using both, which do you prefer? This OCR feature is totally game changing though I wasn’t sure if something like this was even possible it’s almost like magic is this a new feature? Like new within the last few years?
GreatSunJester 1 points 1y ago
They are both good and different. JAWS is more system intrusive, but can sometimes read things better than NVDA. If you need the OCR, then JAWS wins hands down. I think they added it into JAWS in 2018. If you have a scanner, it can import from that and do OCR. I like a non-visual interface for OpenBook.
With JAWS you will get access to better voices, but the ones built into Windows now are not bad. Clearer than eSpeak at least!
I have several clients who I have gotten into using NVDA as a better alternative to Narrator. The cost of JAWS being something some of them need to consider. You might hit up $1 and see what version of JAWS your license is good for.
With JAWS you will get access to better voices, but the ones built into Windows now are not bad. Clearer than eSpeak at least!
I have several clients who I have gotten into using NVDA as a better alternative to Narrator. The cost of JAWS being something some of them need to consider. You might hit up $1 and see what version of JAWS your license is good for.
This nonprofit website is run by volunteers.
Please contribute if you can. Thank you!
Our mission is to provide everyone with access to large-
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
©2023 HumbleCat Inc • HumbleCat is a 501(c)3 nonprofit based in Michigan, USA.