Topics

A call for testing to users of the OCR add-on


Lukasz Golonka
 

Hi all,

i've recently released a new development version of the OCR addon which
can be downloaded from https://github.com/lukaszgo1/nvda-ocr/releases/download/v2.1-dev/ocr-2.1-dev.nvda-addon
While user visible changes are documented in the readme the main change
in this version is the way in which parts of the screen are captured for
recognition and that is where I need your help. If you're using my
add-on currently I would be grateful if you could install the dev
version from the link above and use it for some time.
I've done some testing before release obviously but it is impossible
to account for every application out there, and while the new capturing
implementation makes future maintenance easier I don't want to do that
at the cost of worse recognition results.
Many thanks for your testing!



--
Regards
Lukasz


Robert Doc Wright godfearer
 

Is this suppose to what Jaws does for reading print from a scanner?
******
If we can't look at ourselves, and ask, why? then where does the learning
start?

Family times where there is fun for every ear!
http://stream.wrighthere.net:8000/stream.mp3

Or ask your A device to play Family times on tuneIn
You can also find us on your mobile device install OoTunes and search for
Family times

----- Original Message -----
From: "Lukasz Golonka" <lukasz.golonka@...>
To: <nvda@nvda.groups.io>
Sent: Friday, August 21, 2020 4:01 AM
Subject: [nvda] A call for testing to users of the OCR add-on


Hi all,

i've recently released a new development version of the OCR addon which
can be downloaded from
https://github.com/lukaszgo1/nvda-ocr/releases/download/v2.1-dev/ocr-2.1-dev.nvda-addon
While user visible changes are documented in the readme the main change
in this version is the way in which parts of the screen are captured for
recognition and that is where I need your help. If you're using my
add-on currently I would be grateful if you could install the dev
version from the link above and use it for some time.
I've done some testing before release obviously but it is impossible
to account for every application out there, and while the new capturing
implementation makes future maintenance easier I don't want to do that
at the cost of worse recognition results.
Many thanks for your testing!



--
Regards
Lukasz


Lukasz Golonka
 

Hello,

On Sat, 22 Aug 2020 02:34:42 -0600
"Robert Doc Wright godfearer" <godfearer@...> wrote:

Is this suppose to what Jaws does for reading print from a scanner?

No, this add-on allows to recognize current navigator object with
Tesseract engine similar to what is possible with NVDA built-in OCR but
works regardless of Windows version.

--
Regards
Lukasz


David Ingram
 

How do I get the ocr ad-on? I could not seem to be able to find it. could someone send a direct link. Thank you.

-----Original Message-----
From: Lukasz Golonka <lukasz.golonka@...>
Sent: Aug 23, 2020 4:14 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] A call for testing to users of the OCR add-on

Hello,

On Sat, 22 Aug 2020 02:34:42 -0600
"Robert Doc Wright godfearer" <godfearer@...> wrote:

Is this suppose to what Jaws does for reading print from a scanner?

No, this add-on allows to recognize current navigator object with
Tesseract engine similar to what is possible with NVDA built-in OCR but
works regardless of Windows version.

--
Regards
Lukasz



Lukasz Golonka
 

Hello,

On Fri, 28 Aug 2020 16:14:04 -0500 (GMT-05:00)
"David Ingram" <dingram269@...> wrote:

How do I get the ocr ad-on? I could not seem to be able to find it. could someone send a direct link. Thank you.
It can be downloaded from official add-ons website here:
https://addons.nvda-project.org/addons/ocr.en.html


--
Regards
Lukasz