I am looking for a hotel front desk check in/out function. My perception of how it would appear is that there would be a head shot display able to respond to the common communications topics when checking in a client together with a sensor activated greeting
through a USB controller to detect the approach of the client. The opening sequence when the detection threshold is crossed would be "Welcome to _________ Resort, how may I help you. The rest would follow whatever branch the conversation chose to completion.
My question is, is this the the software I should be looking at?
It's hard to tell from the short description, but it seems that your application may have more to do with speech and less with robotics. The "head shot display" is just a stationary monitor, correct? RDS is good for fusion of many concurrent sensors and
actuators. There is some overhead to enabling this (e.g. wrapping into DSS services, ...) and using it to interface with just a single sensor and do speech may honestly be overly cumbersome.
If this is the extent of your application, you may be better off interfacing directly (with a Kinect or even just a passive infrared motion sensor or whatever) and using Speech Server or SAPI from a regular Windows app.
As for modeling audio dialogs, you might want to look at
VoiceXML which is supported by Speech Server.