Main concepts of the W3C VoiceXML language are:
A session begins once the user begins to interact with a VoiceXML document.
An application is a collection of VoiceXML documents. All the documents in an application share the same application root document.
A form in a VoiceXML document presents information and gathers input from the user.
A menu gives the user a list of choices to select from and transitions to a different dialog or document based on the user's choice.
A link specifies a transition that is common to all dialogs in the scope of the link. When a user input matches the link's grammar, control transfers to the link's destination.
A grammar specifies a list of permissible vocabulary (words and phrases) for the user to select from in order to interact with the VoiceXML application. Each dialog has one or more speech and/or grammars associated with it.
An event is thrown by the VoiceXML platform for a number of reasons, such as when a user does not respond to an input, does not respond correctly, or requests help, etc. The VoiceXML interpreter also throws events in the case that there are semantic errors in the VoiceXML document.