Input & Output

Voice scripts need to communicate to users, and they often do this by telling things to the user through the say element and by defining what they can listen through the expecting element.

A trivial voice script would be:

<app><choice id="launch">
  <say>Good to meet you</say>

The above script gets triggered by saying ‘Hello’, and would respond by saying: ‘Hey. Good to meet you’. When there are multiple say elements, such as in the above example, the output has them all being included serially.

Input Types

Beyond just receiving pre-defined phrases as through the expecting element above, scripts can also receive typed parameters such as a number or a date from a user. In order to do this scripts need to first define the parameter type, for example:

  'name': 'firstName',

Once parameter types have been defined, then scripts can add them to the expecting element parameters, for example:

  <expecting>My name is [[name]]</expecting>
  <say>Hey [[name]]</say>
  <say>Good to meet you</say>

The above script lets the user give their name and greets them accordingly.

Scripts can use one of many supported parameter types including: firstName, lastName, number, date, time, phoneNumber, and phrase.

Output Elements

Beyond the say element, scripts can provide output by:

Furthermore, for output, there is also a potential to also have elements to support displays and other device-specific directives.