Miscellaneous properties
This section describes the miscellaneous properties that Voice Platform supports.
inputmodes
This property controls the input modality. Supported values and associated behavior are:
- dtmf—Enables DTMF-only input.
- voice—Enables voice-only input.
- dtmf voice or voice dtmf (default)—Enables both DTMF and voice input.
One use for this feature includes turning off speech recognition in noisy environments. For example, to disable speech recognition, set the inputmodes property to dtmf. To disable DTMF, set the inputmodes property to voice.
Note: This property does not control the activation of grammars. It only filters methods of user input. For example, setting this property to dtmf does not disable voice-based grammars, it only causes Voice Platform to ignore everything but DTMF input.
nuance.core.tss.outbound.aai
Application-settable property used to specify application-to-application information (aai) that Voice Platform sends in a SIP message.
To send aai in mid-call, the session.setTelephonyProperties() method must be used to specify a new value for this property. To send aai when initiating a transfer, use the aai attribute of the <transfer> element instead.
nuance.grammarlabel
Directs the content of the Nuance-Grammar-Label header. This label is used to identify a particular recognition state among the many recognition states in your Voice Platform application. Voice Platform uses this identifier to group all the call log records for a state.
Nuance recommends that you use the <property> element to populate this property in your application rather than in a context file. Set the property to a meaningful value, such as the name of the current recognition state (for example, GetPizzaSize):
<property name="nuance.grammarlabel" value="GETPIZZASIZE"/>
If you do not set the nuance.grammarlabel property manually in the application, the Voice Browser service generates a value for you. This default is based on a combination of application URL, document URL, form ID, and form item name.
nuance.eventcountergroup
Identifies the set of input events whose counters must be incremented simultaneously. The supported values are noinput, nomatch, and maxspeechtimeout. The default value is an empty set.
It is important to understand that whenever one of the input events within a set is generated, the counters for all input events within that set are incremented as well, for example:
<property name="nuance.eventcountergroup" value="noinput nomatch"/>;
In this example, whenever a noinput event is generated, both the noinput and the nomatch event counters are incremented. The same is true if a nomatch event is generated. Note that the maxspeechtimeout event counter would remain unchanged because it is not specified in the current set of input events.