Topics

OCR 2.0 #addonrelease


Lukasz Golonka
 

Hi all,

After some discussions with NV Access I've decided, with their
permission of course, to maintain the old Tesseract based OCR add-on, as
they don't want to do that going forward. I've created new release
compatible with NVDA 2019.3 and it is available at:
https://github.com/lukaszgo1/nvda-ocr/releases/tag/v2.0
I would appreciate someone updating a link on the website. While at it
I've also updated readme for it so synchronizing its content with the
one present on the add-ons web page along with moving the add-on from
the legacy section would be great.

@Joseph:
I am not sure if OCR is registered in the addon Update but it would be
great if it could be.



--
Regards
Lukasz


 

Hi,
I'll do the following:
* Update documentation based on OCR 2.0 documentation.
* Declare compatibility with 2019.3 on the website.
* Remove OCR from legacy list.
* Update download link for OCR Add-on.
* Register OCR with Add-on Updater.

Of these, everything except the last item will be done this weekend, with OCR add-on registration to follow in Add-on Updater 20.05 (late April).
Cheers,
Joseph

-----Original Message-----
From: nvda-addons@nvda-addons.groups.io <nvda-addons@nvda-addons.groups.io> On Behalf Of Lukasz Golonka
Sent: Saturday, April 18, 2020 10:48 AM
To: nvda-addons@nvda-addons.groups.io
Subject: [nvda-addons] OCR 2.0 #AddonRelease

Hi all,

After some discussions with NV Access I've decided, with their permission of course, to maintain the old Tesseract based OCR add-on, as they don't want to do that going forward. I've created new release compatible with NVDA 2019.3 and it is available at:
https://github.com/lukaszgo1/nvda-ocr/releases/tag/v2.0
I would appreciate someone updating a link on the website. While at it I've also updated readme for it so synchronizing its content with the one present on the add-ons web page along with moving the add-on from the legacy section would be great.

@Joseph:
I am not sure if OCR is registered in the addon Update but it would be great if it could be.



--
Regards
Lukasz


Alberto Buffolino
 

Lukasz Golonka, il 18/04/2020 19.48, ha scritto:
to maintain the old Tesseract based OCR add-on, as
they don't want to do that going forward.
Alberto:
Hi Lukasz,
do you have plan to switch to Tesseract 4.x?
I have been always interested to the project, but the lackness of NVDA internal function for traineddata downloading has stopped me on working forward.
Alberto


Noelia Ruiz
 

Hi, general
ly I
use OCR for Windows 10, but now ```d‘ue `to` `Covid-19 I am working
from home with Windows 7. Thanks for maintaining this.
Would be possible to use the new framework for content recognition, so
that text can be reviewed in browse mode, without using the review
cursor?
Also, let us know if you want to manage requests via GitHub, from this
list or both are right for you.
Chers

2020-04-18 20:16 GMT+02:00, Alberto Buffolino <a.buffolino@...>:

Lukasz Golonka, il 18/04/2020 19.48, ha scritto:
to maintain the old Tesseract based OCR add-on, as
they don't want to do that going forward.
Alberto:
Hi Lukasz,
do you have plan to switch to Tesseract 4.x?
I have been always interested to the project, but the lackness of NVDA
internal function for traineddata downloading has stopped me on working
forward.
Alberto




Lukasz Golonka
 

On Sat, 18 Apr 2020 10:52:40 -0700
"Joseph Lee" <@joslee> wrote:

* Update documentation based on OCR 2.0 documentation.

This has not been done so please update the readme when time permits- at
the moment the info there is outdated.

* Declare compatibility with 2019.3 on the website.
* Remove OCR from legacy list.
* Update download link for OCR Add-on.

Thanks for these.

--
Regards
Lukasz


Lukasz Golonka
 

Hello ALberto,

On Sat, 18 Apr 2020 20:16:07 +0200
"Alberto Buffolino" <a.buffolino@...> wrote:

do you have plan to switch to Tesseract 4.x?
Yes, I do. See https://github.com/lukaszgo1/nvda-ocr/issues/3 for a
tentative plan. While having proper download functionality in NVDA would
be nice it is certainly doable without it.


Lukasz Golonka
 

Hello Noelia,

On Sun, 19 Apr 2020 07:33:46 +0200
"Noelia Ruiz" <nrm1977@...> wrote:

Would be possible to use the new framework for content recognition, so
that text can be reviewed in browse mode, without using the review
cursor?
I've created https://github.com/lukaszgo1/nvda-ocr/issues/4 and we can
discuss this further over there.

Also, let us know if you want to manage requests via GitHub, from this
list or both are right for you.
I'd prefer GitHub issues, however I am following this list so if
something is requested that way i'll probably see it.