Re: supplementary-reader: text reader script

forrest hilton <[email protected]> Sat, 29 Jan 2022 23:19:57 -0600
Newsgroups gmane.comp.kde.devel.accessibility
Message-ID <CAFiqjhkrCv7RbNUJVTFq+VZLO3uwzhp0dFH+Q2J1UTQeiS8uhA@mail.gmail.com>
--00000000000048243005d6c5d296
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Chrys,
I am very interested in working with you. Maybe I could start with adding
mouse based cropping for ocrdesktop.
However since ocr desktop does not info itself provide the screen reading,
I need something additional

My first thought after reviewing some of the "Is there a project that is
jovie successor?" thred is that I should have one or two shortcuts that
reads aloud the middle click keyboard (most recently selected/highlighted
text) then have another shortcut that uses OCR desktop to simulate
selecting text.

As for how to distribute this conveniently, I could just put everything
into a kde shortcuts file. This leaves users on their own for dependencies,
but I don't know if an entire package is warranted for this.

The additional features that I have mentioned are mostly speculation
except for using speech dispatcher, which is critical.

On Sat, Jan 29, 2022 at 11:47 AM Linux A11y <[email protected]> wrote:

> Howdy,
>
> like noted devin, the script looks like an very basic begin to OCRdesktop=
.
>
> I just did an major rewrite. See the full story here:
>
> https://www.patreon.com/posts/ocr-desktop-60895292
>
> https://www.patreon.com/posts/color-analyzing-60988555
>
> https://www.patreon.com/posts/ocrdesktop-3-0-61111548
>
> https://www.patreon.com/posts/ocrdesktop-fun-61131455
>
> https://www.patreon.com/posts/ocrdesktop-ocr-61228691
>
> OCRdesctop features:
> - OCR the whole desktop
> - OCR the current focused window
> - interact with the results like simulate klicks or route the mouse to
> them)
> - OCR files, simple images to multipage PDFs
> - OCR the clipboard content
> - Color analyzing of the recognized text
> - macro functionality for more complex tasks (like open a menu/ submenu
> bevor start OCR process)
> and a lot of other stuff.
>
> It provides an UI to navigate the content, can store the result in
> clipboard or just output to STDOUT.
>
> But its not though to be self voice, this is done by the screen reader.
>
> As you plan to reimplement your functionality, maybe it makes a lot of
> sense to collaborate here and improve OCRdesktop in the given direction.
> It=E2=80=99s written in pure python. Also its already in many distributio=
ns
> repository's. So anyone would benefit.
>
> Would be cool to  join forces.  What do you think?
>
> https://wiki.archlinux.org/title/Ocrdesktop
>
> Cheers chrys
>
>
> ____________________________
>
> See OCRDesktop https://github.com/chrys87/ocrdesktop <https://github.com/=
chrys87/ocrdesktop> for something kinda similar. Orca is a screen reader, i=
t cannot read images though. I don=E2=80=99t know of anything else anywhere=
 remotely similar to this, but I think even blind people could find this us=
eful.
> Devin Praterr.d.t.prater at gmail.com <https://mail.kde.org/mailman/listi=
nfo/kde-accessibility>
> Https://devinprater.flounder.online <https://devinprater.flounder.online/=
>
>
> >* On Jan 23, 2022, at 12:56 PM, forrest hilton <forrestmhilton at gmail.=
com <https://mail.kde.org/mailman/listinfo/kde-accessibility>> wrote:
> *> >* Hello,
> *> >* I have recently written a small script <https://github.com/ForrestH=
ilton/supplementary-reader/ <https://github.com/ForrestHilton/supplementary=
-reader/>> that provides text reader functionality. I will refrain from pas=
ting the entire README, but what makes this different from things discussed=
 in this list is 1) the keyboard shortcuts are available across the desktop=
 2) the program never provides a window or takes up any screen space 3) the=
 program can work on images and pdfs 4) this is a text reader as opposed to=
 a screen reader.
> *> >* I've just today joined this mailing list and only skimmed the archi=
ves. Is there an application that already provides this functionality? Is t=
here a larger project that I can add this functionality to? I might be will=
ing to rewrite the whole thing in another programming language if there is =
a reason.
> *> >* I am aware of a few sloppy things in this script but it works for n=
ow.
> *
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://mail.kde.org/pipermail/kde-accessibility/attachments/2022012=
8/eb559324/attachment.htm>
>
>

--00000000000048243005d6c5d296
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Chrys,<br></div>I am very interested=C2=A0in working =
with you. Maybe I could=C2=A0start with adding mouse based cropping=C2=A0fo=
r ocrdesktop.=C2=A0<div>However since ocr desktop does not info=C2=A0itself=
 provide the screen reading, I need something=C2=A0additional<br><div><br><=
div>My first thought after reviewing some of the &quot;Is there a project t=
hat is jovie successor?&quot; thred is that I should have one or two shortc=
uts that reads aloud the middle click keyboard (most recently selected/high=
lighted text) then have another shortcut that uses OCR desktop to simulate =
selecting text.=C2=A0</div><div><br></div><div>As for how to distribute thi=
s conveniently, I could just put everything into a kde shortcuts file. This=
 leaves users on their=C2=A0own for dependencies, but I don&#39;t know if a=
n entire package is warranted for this.=C2=A0</div></div></div><div><br></d=
iv><div>The additional=C2=A0features=C2=A0that I have mentioned are mostly =
speculation except=C2=A0for using speech=C2=A0dispatcher, which is critical=
.=C2=A0</div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D=
"gmail_attr">On Sat, Jan 29, 2022 at 11:47 AM Linux A11y &lt;<a href=3D"mai=
lto:[email protected]">[email protected]</a>&gt; wrote:<br></div><blo=
ckquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left=
:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"auto">Howdy,<div>=
<br></div><div>like noted devin, the script looks like an very basic begin =
to OCRdesktop.</div><div><br></div><div>I just did an major rewrite. See th=
e full story here:</div><div><br></div><div><a href=3D"https://www.patreon.=
com/posts/ocr-desktop-60895292" target=3D"_blank">https://www.patreon.com/p=
osts/ocr-desktop-60895292</a></div><div><br></div><div><a href=3D"https://w=
ww.patreon.com/posts/color-analyzing-60988555" target=3D"_blank">https://ww=
w.patreon.com/posts/color-analyzing-60988555</a></div><div><br></div><div><=
a href=3D"https://www.patreon.com/posts/ocrdesktop-3-0-61111548" target=3D"=
_blank">https://www.patreon.com/posts/ocrdesktop-3-0-61111548</a></div><div=
><br></div><div><a href=3D"https://www.patreon.com/posts/ocrdesktop-fun-611=
31455" target=3D"_blank">https://www.patreon.com/posts/ocrdesktop-fun-61131=
455</a></div><div><br></div><div><a href=3D"https://www.patreon.com/posts/o=
crdesktop-ocr-61228691" target=3D"_blank">https://www.patreon.com/posts/ocr=
desktop-ocr-61228691</a></div><div><br></div><div>OCRdesctop features:</div=
><div>- OCR the whole desktop</div><div>- OCR the current focused window</d=
iv><div>- interact with the results like simulate klicks or route the mouse=
 to them)</div><div>- OCR files, simple images to multipage PDFs</div><div>=
- OCR the clipboard content</div><div>- Color analyzing of the recognized t=
ext</div><div>- macro functionality for more complex tasks (like open a men=
u/ submenu bevor start OCR process)</div><div>and a lot of other stuff.</di=
v><div><br></div><div>It provides an UI to navigate the content, can store =
the result in clipboard or just output to STDOUT.</div><div><br></div><div>=
But its not though to be self voice, this is done by the screen reader.</di=
v><div><br></div><div>As you plan to reimplement your functionality, maybe =
it makes a lot of sense to collaborate here and improve OCRdesktop in the g=
iven direction.=C2=A0 It=E2=80=99s written in pure python. Also its already=
 in many distributions repository&#39;s. So anyone would benefit.</div><div=
><br></div><div>Would be cool to =C2=A0join forces.=C2=A0 What do you think=
?</div><div><br></div><div><a href=3D"https://wiki.archlinux.org/title/Ocrd=
esktop" target=3D"_blank">https://wiki.archlinux.org/title/Ocrdesktop</a></=
div><div><br></div><div>Cheers chrys</div><div><br><div><br></div>_________=
___________________<br><pre style=3D"color:rgb(0,0,0);background-color:rgb(=
255,255,255)">See OCRDesktop <a href=3D"https://github.com/chrys87/ocrdeskt=
op" target=3D"_blank">https://github.com/chrys87/ocrdesktop</a> &lt;<a href=
=3D"https://github.com/chrys87/ocrdesktop" target=3D"_blank">https://github=
.com/chrys87/ocrdesktop</a>&gt; for something kinda similar. Orca is a scre=
en reader, it cannot read images though. I don=E2=80=99t know of anything e=
lse anywhere remotely similar to this, but I think even blind people could =
find this useful.
Devin Prater
<a href=3D"https://mail.kde.org/mailman/listinfo/kde-accessibility" target=
=3D"_blank">r.d.t.prater at gmail.com</a>

<a href=3D"https://devinprater.flounder.online/" target=3D"_blank">Https://=
devinprater.flounder.online</a>

&gt;<i> On Jan 23, 2022, at 12:56 PM, forrest hilton &lt;<a href=3D"https:/=
/mail.kde.org/mailman/listinfo/kde-accessibility" target=3D"_blank">forrest=
mhilton at gmail.com</a>&gt; wrote:
</i>&gt;<i>=20
</i>&gt;<i> Hello,
</i>&gt;<i>=20
</i>&gt;<i> I have recently written a small script &lt;<a href=3D"https://g=
ithub.com/ForrestHilton/supplementary-reader/" target=3D"_blank">https://gi=
thub.com/ForrestHilton/supplementary-reader/</a>&gt; that provides text rea=
der functionality. I will refrain from pasting the entire README, but what =
makes this different from things discussed in this list is 1) the keyboard =
shortcuts are available across the desktop 2) the program never provides a =
window or takes up any screen space 3) the program can work on images and p=
dfs 4) this is a text reader as opposed to a screen reader.
</i>&gt;<i>=20
</i>&gt;<i> I&#39;ve just today joined this mailing list and only skimmed t=
he archives. Is there an application that already provides this functionali=
ty? Is there a larger project that I can add this functionality to? I might=
 be willing to rewrite the whole thing in another programming language if t=
here is a reason.
</i>&gt;<i>=20
</i>&gt;<i> I am aware of a few sloppy things in this script but it works f=
or now.
</i>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: &lt;<a href=3D"http://mail.kde.org/pipermail/kde-accessibility/attachm=
ents/20220128/eb559324/attachment.htm" target=3D"_blank">http://mail.kde.or=
g/pipermail/kde-accessibility/attachments/20220128/eb559324/attachment.htm<=
/a>&gt;</pre></div></div></blockquote></div>

--00000000000048243005d6c5d296--