I am doing a project around the Kinect. Basically, I have a database of audio files for a number of users.
Out of those users, only one will be standing in front of the Kinect (in its range). The Kinect will be tracking the skeleton, giving him an ID, and also I have managed to display only the tracked skeleton's head. At the same time, I have applied a speech recognition code, so the user will say key words and actions will happen based on those key words. PS: The actions will happen only if the user's voice is found "matched" with one of the audio files in the data base.
Okay, so far so good. All of the above is working relatively okay. Here's the problem: Let's imagine a user standing in front of the Kinect, a "Verified user", at the same time another user walks in, also verified, and says one of the key words. Now when that second user walks in, he will be given a skeleton track ID.
Is there a way that I can pass the track ID of the first skeleton to the speech recognition code so that the actions will happen when the key words are only said by the first user - based on his track ID -?
Thanks.