How to modify amazon lex output voice? - amazon-web-services

Question

I am trying to create a voice bot with aws lex.
In that one of the intents response is "Your incident INC11111111 is closed"(text).
The above response is coming from a lambda function. Please check the code below.
let response = (event, data) => {
let lambda_response = {
"sessionAttributes": {
"incidentNo": event.currentIntent.slots.INCIDENT_NO,
},
"dialogAction": {
"type": "Close",
"fulfillmentState": "Fulfilled",
"message": {
"contentType": "PlainText",
"content": "Hi " + data["User ID"].split('.')[0]+", Your Incident Number " + "INC"+event.currentIntent.slots.INCIDENT_NO+ " is ," + data["Status"]
},
}
};
return lambda_response;
};
Ex Incident No: INC11111111
But the voice output is "your incident INC 1 crore 11 lakhs 11 thousand 1hundered eleven is closed".
What I am expecting is "Your incident INC ONE ONE ONE ONE ONE ONE ONE ONE is closed.
thank you in advance.

Answer 1

You need to utilise SSML(Speech Synthesis Markup Language)
Using SSML tags, you can customize and control aspects of speech, such as pronunciation, volume, and speech rate.
There are a variety of directives that you can use in SSML to pronounce things differently. In your case say-as directive can be useful.
As per the question edit, try these changes
"message": {
"contentType": "SSML",
"content": "<speak> Hi " + data["User ID"].split('.')[0]+", Your Incident Number <say-as interpret-as="characters">" + "INC"+event.currentIntent.slots.INCIDENT_NO+ "</say-as> is ," + data["Status"] +"</speak>"
},
Related reading : Announcing Responses Capability in Amazon Lex and SSML Support in Text Response