Nancy, thanks for looping me in on this discussion. Wendy, it’s a pleasure, 
albeit virtually.

TLDR: this is possible, but may not be a good idea.

Regarding the scanning of text, there are several apps, mainly on IOS, that can 
achieve this. they range from use cases such as scanning a piece of paper to 
scanning printed text on a wall or a sign. There are also magnification apps 
along the same lines that will simply provide some level of magnification. Some 
of these do naïve magnification, but I believe I can also find some examples of 
those that are smart e.g. perform antialiasing and other mathematical 
transformations to the image to enhance the reading of text.

The problem here is one of acquisition. If I’m blind, which I happen to be, 
then I don’t know where the text I want to read is. even if I know the general 
vicinity of the text, or can infer it, (after all, if I feel a sign or wooden 
tablet, it probably has text on it), it’s still not a straight forward task to 
point a camera at it. first off, cameras tend to be on the left most or right 
most of a device, and the natural inclination, especially of low vision/blind 
users is to point the center of the back of the device at the thing they wish 
to take an image of. Next is the lack of preview. So, for example, I won’t know 
if the text I want to read is fully in frame or if I should change orientation 
and so forth. For example, do I need to lift up 6 inches, 9 inches, etc.?

There are ways of mitigating all of the above, so I don’t wish to portray a 
completely hopeless situation. there are some apps that give some sound 
feedback during alignment, others that are simply good at being orientation and 
size invariant, and still others that let you take multiple pictures and stich 
them together to help construct the whole document. But, the end result is that 
these are usually justified when wanting to read an important piece of mail, or 
perhaps answer the most valuable of queries (is this the bottle of my most 
favorite beer or my second most favorite beer), etc. Side note, the beer 
question is most often answered with a bar code scanning app, and/or simply 
having both and taking a stochastic approach to drink pairing.

I realize this may be getting lengthy, but to recap so far, from an on-device 
point of view, there are both magnification and scanning apps with the intent 
for consumption of external text in the environment by a vision-impaired user, 
however, now to the second part of my answer. I’m not sure that this is a good 
idea.

There’s a lot of room for error in this situation as discussed above. Also, 
there’s a high barrier to using this approach for the user. Folks must have the 
apps, or they can be provided, and then the use of them is somewhat complex. 
It’s not hard by any stretch, but it’s also not as easy as “1 2 3” especially 
with retakes and such. Additionally there is an incredibly high variance both 
in the intrauser experience and in the interuser experience, and this may not 
be desired, though I am admittedly totally unfamiliar with your specifics 
requirements, wishes, and desires for the user of course

As I wrap this up, it occurs to me that you may also have been asking about an 
offline approach to this problem, and if you were, I apologize for the above, 
and would say that if you wish to automate the scanning of textual information 
for rendering into speech and enlarged forms, there are some great solutions. 
The first is the use of a document scanner, as these have excellent properties 
for text scanning (assuming whatever you’re scanning is amenable to be ran 
through a document scanner), and then the use of reasonable state of the art 
OCR really does result in a 95% usable situation. the remaining 5% is human 
effort, but can be achieved relatively quickly after some basic training. As 
for rendering into voice (technologies such as text to speech can be used) and 
for enlargement (there’s also a myriad of approaches here depending on 
consumption e.g. paper Vs. screen).

I hope this helps, and please feel free to ask away if I was unclear.

Cheers

Take care,
Sina









President, Prime Access Consulting, Inc.
Twitter: @SinaBahram
Company Website: http://www.pac.bz
Personal Website: http://www.sinabahram.com
Blog: http://blog.sinabahram.com

From: Nancy Proctor [mailto:[email protected]] 
Sent: Wednesday, September 02, 2015 10:53 AM
To: [email protected]
Cc: Sina Bahram <[email protected]>
Subject: Re: Label text scan to audio

Hi Wendy,
You might ask Sina Bahram about his work in this area: Sina Bahram 
<[email protected]> He has been collaborating with a lot of museums on 
accessibility solutions, including the Canadian Museum for Human Rights, the 
Museum of Contemporary Art in Chicago, and the Baltimore Museum of Art, as well 
as on the digita11y.org access app project.

I look forward to hearing what you find out!
Nancy
-- 
Nancy Proctor, PhD
Deputy Director for Digital Experience and Communications, BMA
[email protected]
o: +1 (443) 573-1596
m: +1 (301) 642-6257
@NancyProctor

Baltimore Museum of Art 
10 Art Museum Drive
Baltimore, MD 21218
+1 (443) 573-1700
http://artBMA.org
@artBMA

Date: Thu, 27 Aug 2015 14:46:26 +0000
From: Wendy Sporleder <[email protected]>
To: 'Museum Computer Network Listserv' <[email protected]>
Subject: [MCN-L] Label text scan to audio
Message-ID:
<blupr04mb006e2073838a37c622cfa4e96...@blupr04mb006.namprd04.prod.outlook
.com>
Content-Type: text/plain; charset="us-ascii"

To assist visually impaired visitors, is anyone having success with text
scan to audio/large font text using an app or similar?

Wendy Sporleder
Database Administrator
Saint Louis Art Museum
One Fine Arts Drive, Forest Park
Saint Louis, MO.  63110
314.655.5318
[email protected]<mailto:[email protected]>,

_______________________________________________
You are currently subscribed to mcn-l, the listserv of the Museum Computer 
Network (http://www.mcn.edu)

To post to this list, send messages to: [email protected]

To unsubscribe or change mcn-l delivery options visit:
http://mcn.edu/mailman/listinfo/mcn-l

The MCN-L archives can be found at:
http://www.mail-archive.com/[email protected]/

Reply via email to