…be programmed to let you play online poker, making voice, checks, raises, folds, etc., instead of using the mouse and/or keyboard?
I mean programming it ,not by a computer whiz (such as myself) but, by someone with considerably less talent (such as Shagnasty and other slow Dopers).
Anything you can do on your computer can be simulated by a program. Whether or not you’d be able to pull it off with your skills depends on the complexity of the task at hand.
Why don’t you give it a shot and let us know? One person’s average guy is another person’s super genius. There’s no telling what your average average guy will be able to accomplish if he sets his mind to it.
Generally, Dragon and similar speech recognition programs work by simulating keyboard input. That is, they ‘listen’ to your speech, convert that to text, and then feed that into the computer as if you had typed that text at the keyboard.
I haven’t played online poker, but most PC games depend a lot on use of the mouse. Often the programmer fails to include appropriate keystroke commands for all game activities. (A blind friend, who uses a screen reader, frequently complains about programs that do this.)
So the first thing I would do is see if you can play this game using ONLY the keyboard. If so, then there is a fair chance that you could make it work with a speech recognition program, and you can look into it further. (I think there are free trial periods for some such programs.)
I haven’t used Dragon, nor do I use Windows (IIRC, Dragon is a Windows only program), so I can’t comment on them specifically. However, my dissertation work deals with human-robot interaction, and I’ve used the (open source, free) Sphinx speech recognition program a whole lot (and one or two others).
Anything that takes text as input can pretty easily be speech-controlled; you just have to know where to pipe it (i.e., where to obtain the speech program’s output and where to redirect it as the poker program’s input). The issue here, ISTM, is mostly becoming familiar with the poker client’s API. If you’re entering text, it should be relatively easy to redirect recognized text to the client’s input. The mouse would be a little trickier (I think), as it’s at the OS level.