Detailed Implementation Notes

This section includes additional detail for the items contained in the VoiceXML Summary.

<audio> Element

The VoiceXML interpreter processes audio content so that only text is sent to the TTS engine; pre-recorded audio is played by the telephony hardware. The VoiceXML interpreter:

u Extracts the audio content from a SSML container

u Sends content before and after <audio> content to the TTS processor

u Processes audio content (for example, a referenced WAV file) using the associated telephony extension point for the call

In the following example, the VoiceXML interpreter will extract the <audio expr="greeting"/> so that it can be played by the telephony hardware. The remaining content is processed through the TTS engine.

<data> Element

By default, the VoiceXML interpreter will allow a voice application to access returned XML content when an access-control processing instruction is not specified.

A Java system property – vocalos.vxml.data.access_control.allow – can be set to "false" to override this default behavior. This system property is not delivered (the VoiceXML interpreter behaves as if it were set to "true").

To add the parameter, edit the vocalos.conf (Windows) file and add the -D system parameter as shown in the following example.

<log> Element

To log information using the Apache Software Foundation log4j facility, insert the optional label attribute using the following syntax:

Note: Check the log4j.xml file to make sure the logging level you specify with the label attribute is defined for the category.

Valid logging levels – set with "priority" in the above example – are:

u DEBUG (default)

u INFO

u WARN

u ERROR

u FATAL

For example, the following attribute will log to the log4j category, "sample," using the debug level.

<meta> Element

The following meta properties are supported.