0

I'm using Microsoft's UCMA 4.0 SDK to build a basic VXML IVR system. When I have a normal prompt using the default system voice, you can 'barge-in' by pressing any DTMF key:

<prompt>Your credit card number is 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4</prompt>

However I'd like to use the customers own voice. To do this i build up an array of individual audio prompts, and play them back using a loop:

<prompt>
    <foreach item="i" array="BuildCardPrompt()">
        <audio expr="i"/>
    </foreach>
</prompt>

Problem is that now I am unable to 'barge-in' and interrupt the loop. Any idea how to solve this issue?

markv
  • 275
  • 1
  • 5

1 Answers1

0

The VoiceXML spec indicates that barge-in should work with a foreach loop. Might be a platform issue and I do not have the Microsoft platform to test it out. You could try wrapping the audio element in a prompt element like this and see if it makes a difference.

  <foreach item="i" array="BuildCardPrompt()">
    <prompt>
      <audio expr="i"/>
      <break/>
    </prompt>
  </foreach>

Another alternative is to dynamically create the VoiceXML that contains the sequential series of prompts.

Kevin Junghans
  • 17,475
  • 4
  • 45
  • 62