Hi all
I am trying to write an application that will respond to voice commands and
then give feedback with voice using text to speech. This is being written in
.NET 3.x using C#
I got the TTS stuff working but am having a bear of a time with the speech
recognition.
There does not seem to be much written about it and what is written often is
just plain wrong. Or I am just stupid. Either way, I am having problems.
The xml grammar file is blowing my mind.
I can get very simple one word or a hard coded sentence to be recognized but
I know the app should be much more flexible than that. What I would like to
do is have a group of 4 commands that can be recognized as single words OR as
words embedded in common phrases. Basically, I am only interested in the
single command but would like the user to be able to ask it in several
different ways.
Once a command is chosen, I would like to then have another set of commands
be available and perhaps more commands under each of them in a tree type
structure. However, the user should only be able to issue commands that are
current at a particular node level of the tree. None below, above, or
siblings should be able to be issued.
I know that this can be done with an xml based grammar file but I am
floundering at the moment.
Could someone please upload a simple example of how I could implement such a
thing in the xml grammar? Or point me to a resource that is more clear than
the current dot net documentation?
many thanks,
clay