Volume 1, Issue 2 - Feb. 2001
   
   
 

VoiceXML: Where Speech Meets the Web

By Rob Marchand

(Continued from Part 1)

For a better example of how elements and containers nest, let's have a look at our third example from last month:

Example 3: Advanced Hello World (excerpt)
 
<?xml version="1.0"?>
<vxml version="1.0">
      <!--Example 3 for VoiceXML Review -->
      <form>
            <block>
                  <prompt>
                        <audio
src="http://www.voicexml.org/audio/helloworld.wav">
                        Hello, World!
                        </audio>
                  <prompt>
            </block>

            <field name="greeting">

                  <prompt>
                        What say you?
                  </prompt>

                  <grammar>
                        hello | howdy | greetings | hey | password
                  </grammar>

                  <help>
                        You can say hello, howdy, greetings, hey, or
                        password
                  </help>

                  <filled>
                        You said <value expr="greeting"/>
                  </filled>

            </field>

            <block>
                  <!--Decide whether to continue talking to this caller -->
                  <submit next="http://www.voicexml.org/
                        cgi-bin/friend_or_foe.cgi"/>
            </block>

      </form>
</vxml>


And here is the stylized version:

Here we see the structure of a sample <field> element, used to collect input from the user. This field element has a number of children:

  • prompt - The announcement to be presented to the use to convince them to say something.

  • grammar - a definition of what is considered legal speech input to this field item. In this case, a set of alternative greetings is acceptable.

  • filled - a description of the actions to undertake when the field is considered 'filled'. A field is declared filled (in this case) when the use has said something matching the grammar: 'hello', 'howdy', or one of the other grammar alternatives.

  • Help Event - A description of the actions to undertake when the user requests help. In this example, merely some text indicating what is acceptable (this is read to the user).

It's important to understand this structure, particularly when starting out, as it will enable you to avoid headaches caused by improperly formed documents. It's a good idea to find and use an XML editor and/or validator that can take advantage of the VoiceXML DTD to ensure your pages are 'well-formed'.

What's Next?

Next month, we're going to talk about the features in VoiceXML for building bigger applications, and how you can structure such an application. In the meantime, you might want to have a look at these resources provided by the VoiceXML Forum:

A number of VoiceXML Forum Members provide access to developer sites and tool kits that will allow you to try out VoiceXML for yourself. A few of these are:

back to the top

 

Copyright © 2001 VoiceXML Forum. All rights reserved.
The VoiceXML Forum is a program of the
IEEE Industry Standards and Technology Organization (IEEE-ISTO).