VoiceXML application properties allow you to set platform defaults for a session and alter platform behavior when the application executes. Unsupported properties are allowed but ignored at execution. Properties follow the standard VoiceXML scoping rules.
To use a property in a VoiceXML application, specify the value attribute for the <property> tag. For example, suppose you want to have an application that takes advantage of a fast caching policy, so that your application (VoiceXML documents, grammars, scripts, and so forth) load quickly. However, during development you need your application to use a safe caching policy so that you can develop and debug. During development, set the following properties in the application root:
<vxml version="2.1">
<property name="documentmaxage" value="0"/>
<property name="audiomaxage" value="0"/>
<property name="grammarmaxage" value="0"/>
<property name="scriptmaxage" value="0"/>
...
</vxml>
When you deploy the application, either change the property values from 0 to non-zero values, or remove the properties.
Note: A property is scoped to the level where it is specified. For example, a property specified at the application level provides the defaults for all loaded documents. A property at the form level overrides properties set previously at the application or document level.
The VoiceXML Interpreter supports the following types of properties:
The following table describes the VoiceXML Interpreter's DTMF properties:
Property |
Data Type |
Default |
Description |
interdigittimeout |
Number |
3.0s |
Timeout value between DTMF occurrences. |
termchar |
0-9 |
# |
Terminating DTMF character for DTMF input recognition. Using the blank option ("") allows you to specify all DTMF characters as input items, including the pound key (#). |
termtimeout |
Number |
0s |
Termination timeout for DTMF occurrences. |
The following table describes the VoiceXML Interpreter's fetch properties.
Property |
Data Type |
Default |
Description |
audiofetchhint |
{safe | prefetch} |
safe |
Not currently supported. |
audiomaxage |
Number |
NA |
Maximum acceptable age of cached audio resources, in seconds. |
audiomaxstale |
Number |
NA |
Maximum acceptable staleness of expired, cached audio resources, in ms. Specifying an audiomaxstale value lets a VoiceXML document use audio resources contained in the cache for a given period after the resources have expired. |
datafetchhint |
{safe | prefetch} |
safe |
Not currently supported. |
datamaxage |
Number |
NA |
Maximum acceptable age of cached XML data, in seconds. |
datamaxstale |
Number |
NA |
Maximum acceptable staleness of expired cached XML documents, in seconds. Specifying a datamaxstale value lets a VoiceXML document use XML data contained in the cache for a given period after the data has expired. |
documentfetchhint |
{safe | prefetch} |
safe |
Not currently supported. |
documentmaxage |
Number |
NA |
Maximum acceptable age of cached documents, in seconds. |
documentmaxstale |
Number |
NA |
Maximum acceptable staleness of expired cached documents, in seconds. Specifying a documentmaxstale value lets a VoiceXML document use VoiceXML documents contained in the cache, for a given period after the documents have expired. |
fetchaudio |
Number |
NA |
URI of the audio to play while waiting for a document to be fetched. |
fetchaudiodelay |
Number |
0s |
Amount of time to wait at the start of a fetch delay before playing the fetchaudio source. This is useful if a fetch delay is short, when it may be preferable not to start playing fetchaudio that will be cut-off almost immediately. |
fetchaudiominimum |
Number |
2s |
Minimum time interval to play a fetchaudio source, once started, even if the fetch result arrives in the meantime. This is useful when you do not want the fetchaudio heard by a user to be stopped too quickly. |
fetchtimeout |
Number |
10s |
Amount of time to wait before a timeout event is raised on HTTP fetches. |
grammarfetchhint |
{safe | prefetch} |
safe |
Not currently supported. |
grammarmaxage |
Number |
NA |
Maximum acceptable age of cached grammars, in seconds. |
grammarmaxstale |
Number |
NA |
Maximum acceptable staleness of expired cached grammars, in seconds. Specifying a grammarmaxstale value lets a VoiceXML document use grammar resources contained in the cache, for a given period after the resources have expired. |
objectfetchint |
{safe | prefetch} |
safe |
Not currently supported. |
objectmaxage |
Number |
NA |
Maximum acceptable age of cached objects, in seconds. |
objectmaxstale |
Number |
NA |
Maximum acceptable staleness of expired cached objects, in seconds. Specifying an objectmaxstale value lets a VoiceXML document use objects contained in the cache for a given period after the objects have expired. |
scriptfetchhint |
{safe | prefetch} |
safe |
Currently not available. |
scriptmaxage |
Number |
NA |
Maximum acceptable age of cached scripts, in seconds. |
scriptmaxstale |
Number |
NA |
Maximum acceptable staleness of expired cached scripts, in seconds. Specifying a scriptmaxstale value lets a VoiceXML document use script resources contained in the cache for a given period after the resources have expired. |
The following table lists the VoiceXML Interpreter's prompt properties:
Property |
Data Type |
Default |
Description |
bargein |
boolean |
yes |
Allows or disallows prompt bargein. |
bargeintype |
{speech | hotword} |
speech |
Type of barge-in performed in response to voice input. |
timeout |
Number |
3.4s |
Elapsed silent time before a noinput event is raised. |
The following table lists the VoiceXML Interpreter's recognition properties:
Property |
Data Type |
Default |
Description |
completetimeout |
0.2s - 10s |
0.25s |
Length of silence after speech before a result is either accepted or rejected with a nomatch event. This value applies when the speech prior to silence is a complete match of an active grammar. |
confidencelevel |
0.0 - 1.0 |
0.5 |
Threshold of the speech recognition confidence level. The VoiceXML Interpreter rejects user inputs if the confidence level is below this threshold. Valid values range from 0.0 to 1.0. |
incompletetimeout |
0.2s - 10s |
0.75s |
Length of silence after speech before a result is finalized. This value applies when the speech prior to silence is either:
In the first case, when the timeout is triggered, the VoiceXML Interpreter rejects the partial result and raises a nomatch event. In the second case, the VoiceXML Interpreter accepts the speech. |
inputmodes |
{dtmf | dtmf voice | voice} |
dtmf voice |
Space separated list defining the type of inputs allowed. Valid values:
|
maxspeechtimeout |
0s - 600s |
15s |
Maximum duration of user speech. In VoiceXML 2.0, if this time elapses before the user stops speaking, a maxspeechtimeout event is raised. Use 0 for no limit. |
maxnbest |
1 - 500 |
1 |
Maximum number of results returned by the recognizer. Also represents the maximum size of the application.lastresult$[i] array. |
recordutterance |
boolean |
no |
Indicates whether to enable or disable utterance recording during recognition.
When utterance recording is enabled, the following variables contain recording values:
If a form item is filled, the corresponding form item shadow variables also are set to these values. |
recordutterancetype |
{audio/x-wav | video/3gpp} |
audio/x-wav |
Media type to use for utterances recorded during recognition. |
sensitivity |
0.0 - 1.0 |
0.5 |
Sensitivity level to input. Valid values range from 0.0 (least sensitive) to 1.0 (most sensitive). |
speedvsaccuracy |
0.0 - 1.0 |
0.5 |
Provides information to the recognizer of the desired balance between speed and accuracy. Values range from 0.0 (fastest recognition) to 1.0 (highest accuracy). |
universals |
{cancel | exit | help | none} |
none |
Space-separated list of the universal commands to activate; for example: "help" "cancel" "exit" |