Description

Returned when NlpIntentModel was specificed in the ListeningOptions, it contains the results of the NLP Intent model classification on the last sentence. NlpIntentResponse will only run on complete sentences (isFinalTranscription = true).

Example

//@input Asset.VoiceMLModule vmlModule

var parseAdditionalParams = function(additionalParams) {
if (additionalParams.length == 0) return "";
var retParams = "Additional params";
for (var i = 0; i < additionalParams.length; i++) {
retParams += " Key: " + additionalParams[i].key + " Value: " + additionalParams[i].value;
}
return retParams;
}

var parseStatusCode = function(status) {
var code = "";
switch(status.code) {
case VoiceMLModule.NlpResponsesStatusCodes.OK:
code = "OK";
break;
case VoiceMLModule.NlpResponsesStatusCodes.ERROR:
code = "ERROR";
break;
}
return "\nStatus Code: " + code + " Description: " + status.description;
}

var onUpdateListeningEventHandler = function(eventArgs) {
var intentResponses = eventArgs.getIntentResponses();
var nlpResponseText = "";

for (var iIterator = 0; iIterator < intentResponses.length; iIterator++) {
intentResponse = intentResponses[iIterator];
nlpResponseText += "Intent Response: " + intentResponse.modelName + "\n intent: " + intentResponse.intent;
nlpResponseText += parseAdditionalParams(intentResponse.additionalParams);
nlpResponseText += parseStatusCode(intentResponse.status);
nlpResponseText += "\n\n";
}
}

script.vmlModule.onListeningUpdate.add(onUpdateListeningEventHandler);
interface NlpIntentResponse {
    additionalParams: AdditionalParam[];
    intent: string;
    modelName: string;
    status: NlpResponseStatus;
    getTypeName(): string;
    isOfType(type): boolean;
    isSame(other): boolean;
}

Hierarchy (view full)

Properties

additionalParams: AdditionalParam[]

Description

Additional params to send in the response.

intent: string

Description

The classification result of the NLP intent model for the sentence. For example for the sentence "show me the next item" with the model VOICE_ENABLED_UI, the result will be next for VOICE_ENABLED_UI the following intents are valid: next, back, left, right, up, down, first, second, third, fourth, fifth, sixth, seventh, eighth, ninth, tenth. In addition, the following values could be returned:

#SNAP_ERROR_INCONCLUSIVE: two or more intents detected. #SNAP_ERROR_INDECISIVE: no intent detected. #SNAP_ERROR_NONVERBAL: no transcribable human voice was detected. #SNAP_ERROR_SILENCE: silence was detected.

Anything starting with #SNAP_ERROR_: Errors that are not currently defined in this documentation and should be ignored.

modelName: string

Description

Name of the model that sent the response. In current studio release, you can expect the result to contain either VOICE_ENABLED_UI like described in NlpModel. If using enableSystemCommands(), SYSTEM_COMMANDS will be returned as well.

Description

The status of the NLP response. in case of successful result, the value should be VoiceMLModule.NlpResponsesStatusCodes.OK, in case of failure, the value should be VoiceMLModule.NlpResponsesStatusCodes.ERROR.

Methods

  • Parameters

    • type: string

    Returns boolean

    Description

    Returns true if the object matches or derives from the passed in type.

  • Parameters

    Returns boolean

    Description

    Returns true if this object is the same as other. Useful for checking if two references point to the same thing.

Generated using TypeDoc