Hi Jim,

I see. Well, I know back when I was using Jaws for Windows. This was
perhaps some 10 years ago I could set Jaws to "say all" and it would
read everything in the command prompt window. I was using something
like frotz and could load up a game like Arthor and it did a decent
job of reading text automatically as long as "say all" was enabled.
Still I take your point. SAPI and prerecorded speech has given
everyone a truly screen reader independant design that doesn't require
a screen reader at all let alone weather it has good command
prompt/terminal support.

For instance, I myself use a screen reader called Orca on Linux. Most
of the time it does ok with gnome-terminal, something similar to the
Windows command prompt window, and I can run text-based programs in
it. Unfortunately, Orca is sort of hit or miss with these games etc.
You might be using Scare and it will fail to read all of the text on
the screen, and the mouse cursor might not be able to read the text
either. For that reason alone when I play Adrift, Tads, or Inform
based text adventures under Linux I often exit the graphical user
environment all together, log into the shell itself, and use a screen
reader called Speakup that was specifically designed for text-based
programs. While it doesn't necessarily read the text automatically I
can review the screen and see what is going on. So I do understand and
know where you are coming from. From that point of view it would be
nice just to use speech-dispatcher and shoot the speech out to one of
the Cepstral voices, ESpeak, Festival, AT&T Mike, etc rather than
fooling with reviewing the screen.

On the other hand what I really wanted to do is write a game like
Piledriver. I.E. just a simple text -based wrestling game. Nothing too
special or complicated which is what we get if we have to include
SAPI, unicode support, and all that other crap that goes with Windows
API programming. Unfortunately, for a C++ programmer you cant just
send an ascii text string like
"hello world!"
to SAPI 5. Instead the Microsoft Windows API forces you to convert
that text string to unicode, and then you can pass it to the Speak()
function.  That's something of a pain which is one reason I haven't
used SAPI in a while. I pretty much have to write my own convert
functions to convert a single word or line of text to and from unicode
complecating the source code. Plus making it totally operating system

I guess the best thing to do here would be to use something like
Scansoft Tom and just record everything from scratch. Then, I could
use LibSDL, which is totally cross-platform, for sounds, music,
sspeech, and input. I'm already doing this for STFC  so there is
nothing barring me from doing this with a wrestling game as well.


On 4/10/11, Jim Kitchen <j...@kitchensinc.net> wrote:
> Hi Thomas,
> Other than web based, I do not know of a single blind accessible game that
> puts text on the screen that gets read automatically by all windows screen
> readers.  It has never worked that way.  That is why we have always used
> recorded speech and or sapi5.

Gamers mailing list __ Gamers@audyssey.org
If you want to leave the list, send E-mail to gamers-unsubscr...@audyssey.org.
You can make changes or update your subscription via the web, at
All messages are archived and can be searched and read at
If you have any questions or concerns regarding the management of the list,
please send E-mail to gamers-ow...@audyssey.org.

Reply via email to