i am developing a tool, which makes it possible for users to control their full working copy of DNS using a Philips SpeechMike. SpeechMike is a multifunctional USB device, mainly used to dictate into a dictation software. it has some typical buttons on it like Record, Stop, Play, FFWD, FRWD and some extra function buttons.
so if the user presses Record on the SpeechMike, dragon switches the Mic to "on", if the user presses the Play button on the SpeechMike dragon starts the playback, etc.
it´s easy to utilize the dragon SDK to turn the mic on. i have some troubles with starting the playback though. all I want to do is start the playback on text dictated into any window (Word, Notepad, …), just like if you were using the DNS menu to start it, or if you were saying "Play back" into the mic.
it seems that it´s quite much harder to use the SDK to start playback. there are two different SDK objects, which provide a Playback function: the Dictation-edit object and the Custom dictation object.
I can´t use the Dictation-edit object, because it has to be attached to one edit control. Ok, I can enable the compatibility modules, so at least I can dictate into any window, but still the playback won´t work.
So I decided to use the Custom dictation object. I register one for each window, I want to dictate to. the problem here is, that i must write my own code to keep the internal dictation buffer and the associated window synchronized. this is lots of work, since I would have to be able to write/read text to/from all windows/programs, I can dictate to.
is it really that complicated? all I want to do is tell the running DNS to start playback! I just want DNS to care for everything, i.e. where is the cursor, what should be played back etc.
virtual keypress events would be an option, but there will always be some problems with setting the focus…

Please help!
Robert
p.s.: I want to note, that I really appreciate all the work, Chuck Runquist is doing just by answering so much questions. Where do you take the time?