With regard to recently announced internet radio products and in general,
here are a few comments about how hardware gadgets can and can
not be made accessible through speech output.  I address only
speech output for the purposes of substituting for reading a
visual display.  Controlling things, via mice, buttons, pointing
devices speech input etc. is an almost separate issue.

First, I am a totally blind rehabilitation engineer at
Smith-Kettlewell Rehab Engineering Research center and have been
bashing my head against these walls for almost 19 years with
extremely variable success.

It "APPEARS" from the description of the aforementioned internet
radio, that one looks at the screen and uses a knob to "tune" in
selected areas of interest and particular "stations."

It seems very likely that one can not "tune" as with a radio, and
hear stations one passes by.  After all as you select a station,
the device has to "connect" to it just as does your browser so
that takes an intent to connect and some noticable time.

So you would probably have to see it to run it.

There is no magic bullet "speech chip" that can read displays.

The fact that you can buy a talking clock for $10.00 is not
relivant, the clock knows what it's supposed to talk about, and devices
that were'nt designed to talk do not know they should.
As for displays, First there are many different kinds of displays
and many many different ways of putting information on them.  In
the case of this device, it is likely a video signal, much like
the one that goes between your computer and its monitor.

Many bright people have tried and are trying to make a device
that will look at those kinds of video signals and "read" them.
That problem makes reading machines and there tasks look easy,
and to work interactively, it'd have to be very fast.

The solution, if there is one, is almost always found in
modifying the software that runs the device in question to get the
information you need to interact with.  You
have then to add the "speech chip" or speech synthesis software
to actually talk.

How complicated the software modifications may be depends on how
complicated the interaction between user and screen is.  windows
being a near worst case example of unnecessary and hideous
complexity, most will agree that screen readers only "sort of"
work in that environment even now.

In the case of a simpler device wherein you are merely moving
through 1 or 2 dimensional lists, the problem might be rather
easier.

If, and only if, the manufacturer could be persuaded to care,
they might find it not too difficult to have the selected portion
of the display dumped as text either through a serial port, or
more conveniently to built-in speech synthesis software.

If memory serves at least one of the popular speech synthesis
packages is available for linux, and the radio's advertising says
they are running a linux based operating system.

Without further information, that's about all that I can say with
any probability of guessing rite.

I was told recently by a salesman of access equipment that soon
"everything will have an I.P. address" then it will all be fine."
but then his company offered a catalog in PDF on this list. <G>
I am not holding my breath.

However with a device like that radio, as it uses a common
operating system, and its interaction needs may not be too
complex, I can imagine an access solution if the maker cares.

So if you see a need for this device, (Personally, the idea of
4000 radio stations bores me beyond tears,) write the company  or email
them and see if we can get there attention.

And, any linux nerds out there be ready with sources of text to
speech software and other ideas if they bite, or is it byte.

By the way, we here at Smith-Kettlewell have a high powered
machine vision scientist working on a machine that would use an
inexpensive "web camera" and a computer to
"read" things with visual displays.  Although there is some basic
progress, it is a very difficult task, as I said before it makes
reading machines look simple, and no it is almost an entirely
different problem from that faced by reading machines so you can
not just use "OCR" software.  Recognizing the characters is
simple, you gotta "find" em first.

With this project,We hope to approach the huge problem of
consumer products with displays, appliances and so on.  We don't
think that being able to get maybe one of a particular thing that
talks and costs twice as much, is more than a good start at
having an accessible life.

Hope this helped a bit.
Screen reader sales men please don't bother me.

Tom Fowle
Embedded Systems Developer/ Rehab engineer
Rehabilitation Engineering Research Center
The Smith-Kettlewell Eye Research Institute
2318 Fillmore St.
San Francisco, CA 94115
415-345-2123 )
Internet [log in to unmask]


VICUG-L is the Visually Impaired Computer User Group List.
To join or leave the list, send a message to
[log in to unmask]  In the body of the message, simply type
"subscribe vicug-l" or "unsubscribe vicug-l" without the quotations.
 VICUG-L is archived on the World Wide Web at
http://maelstrom.stjohns.edu/archives/vicug-l.html