Articulation scores have been experimentally obtained as functions of varying word content, bandwidth, audio signal-to-noise ratio and the experience of the talkers and listeners involved.
Similarly, when recognizing a talker, all the memory traces of utterances produced by that talker are activated and the talker's identity is determined.
The talker can explicitly tear down a stream by de-registering the talker advertise, and a listener can disconnect by de-registering the listener ready.