VoiceXML Review - Columns - Speak & Listen

In this monthly column, an industry expert will answer common questions about VoiceXML and related technologies. Readers are encouraged to submit questions about VoiceXML, including development, voice-user interface design, and speech technology in general, or how VoiceXML is being used commercially in the marketplace. If you have a question about VoiceXML, e-mail it to and be sure to read future issues of VoiceXML Review for the answer.

This month we received more excellent questions from our readers. It's great to see momentum begin to develop here, and I look forward to the point where too many questions are coming in each month to publish answers to every one!

Q: I'm writing a script using the <transfer> tag. I'm trying to place the call dynamically using a number that the user speaks. I'm not having any luck. Any advice?

A: The attribute "detextpr" is available on <transfer>, which lets you use a JavaScript expression to set the destination phone number. For example:

<vxml version="1.0">
  <form name="transfer">
    <!-- Set a form-scope variable to be some phone number.
         This could have been collected using a previous form or field -->      
    <var name="mydest" expr="'8005558355'"/>
    <block>
      <audio>Ready to attempt call transfer</audio>
    </block>
    <!-- Now transfer using the destexpr attribute to use a JS expression -->
    <transfer destexpr="mydur">
      <filled>
        <audio>say something here</audio>
      </filled>
    </transfer>
    <block>
      <audio>welcome back</audio>
    </block>
  </form>
</vxml>

A: Authentication can be done in many ways, just like on the Web. None of these are specific to VoiceXML as a language (just as on the Web they're not specific to HTML). Typically, there are two styles used--either strictly numeric ids/passwords that customers can say or type in on their handset, which have no recognition issues, or a new technology variously called "voice printing" or "voice verification" , which is a biometric authentication scheme that prompts the user to speak a generated phrase, which is compared to a previously recorded example of their speech. For more information on this kind of techology, visit the Nuance (http://www.nuance.com) and SpeechWorks (http://www.speechworks.com) web sites. Regarding integrating sites that offer both voice and Web based access, the integration is done on the back end, and the site can offer either the same id/password for both (e.g. a numeric one), or a distinct one for voice vs. web, which are linked to the same account on the back end. In almost all cases, SSL (Secure Sockets Layer) technology is used to secure communications between Web servers and VoiceXML interpreters--exactly like it is used in traditional HTML applications between Web servers and Web browsers.

Answers to Your Questions About VoiceXML

By Jeff Kunins

Q: I'm writing a script using the <transfer> tag. I'm trying to place the call dynamically using a number that the user speaks. I'm not having any luck. Any advice?