Teaching Alexa New Skills

Part 1: The Amazon Echo

Why Program for Echo?

Allows the developer to work in a voice-controlled environment without needing to build a speech recognition engine of their own.

Amazon's service sends your app parsed responses in string form.

The Echo App Workflow

Teaching Alexa New Skills

Part 2: AWS Lambda & Echo's Dev Workflow

What is Lambda?

  • AWS Lambda is a compute service that runs developers' code in response to events and automatically manages the compute resources for them, making it easy to build applications that respond quickly to new information.
     
  • Supports Node.js, Python, or Java.
     
  • Requires no external hosting (though if you want to host your own service, Alexa can use that too).

Five Simple Steps

a. Settle on an Idea

b. Create your Intent Schema

c. Provide Alexa some Sample Utterances

d. Write your Lambda Function

e. Create your skill

Teaching Alexa New Skills

Part 3: Echo's Echo

3a. The Idea

Why doesn't a device named 'Echo' have the ability to echo responses?

3b. The Intent Schema

Tell Alexa the types of commands allowed in your app

{
  "intents": [
    {
      "intent": "EchoIntent",
      "slots": [
      {
        "name" : "Echo",
        "type": "AMAZON.LITERAL"
      }
     ]
    },
    {
      "intent": "AMAZON.HelpIntent"
    },
    {
      "intent": "AMAZON.StopIntent"
    }
  ]
}

3c. Sample Utterances

Tell Alexa the form of commands and interactions

EchoIntent {echoa|Echo} {echob|Echo}

But wait? Why wouldn't this work...

EchoIntent {only_echo|Echo}

3d.1 'Main' Function

exports.handler = function (event, context) {
    try {
        if (event.session.new) {
            onSessionStarted({requestId: event.request.requestId}, event.session);
        }
        if (event.request.type === "LaunchRequest") {

            // call onLaunch function

        } else if (event.request.type === "IntentRequest") {

            // call onIntent function

        } else if (event.request.type === "SessionEndedRequest") {

            onSessionEnded(event.request, event.session); // Handle session end
            context.succeed();
        }
    } catch (e) {
        context.fail("Exception: " + e);
    }
};

3d.2 Intent Router

//Called when the user specifies an intent for this skill.

function onIntent(intentRequest, session, callback) {

    var intent = intentRequest.intent,
        intentName = intentRequest.intent.name;

    // Dispatch to your skill's intent handlers
    if ("EchoIntent" === intentName) {

        echoReply(intent, session, callback); // Handle the EchoIntent

    } else if ("AMAZON.HelpIntent" === intentName) {

        getWelcomeResponse(callback); // Handle the HelpIntent

    } else if ("AMAZON.StopIntent" === intentName) {

        getStopResponse(callback); // Handle the StopIntent

    } else {

        throw "Invalid intent";

    }
}

3d.3 Welcome Handler

//function getStopResponse(callback)

function getWelcomeResponse(callback) {
    // If we wanted to initialize the session to have some attributes
    // we could add those here.
    var sessionAttributes = {};

    var cardTitle = "Welcome";

    var speechOutput = "Entering echo mode. " +
        "I will now repeat everything you say until I hear 'stop'" +
        ", or it stays quiet for 5 seconds.";

    // If the user either does not reply to the welcome message or
    // says something that is not understood, they will be prompted
    // again with this text.
    var repromptText = "";

    var shouldEndSession = false;

    callback(sessionAttributes,
        buildSpeechletResponse(cardTitle, speechOutput, 
            repromptText, shouldEndSession));
}

3d.4 Echo Handler

function echoReply(intent, session, callback) {

    var cardTitle = intent.name;

    var wordsToRepeat = intent.slots.Echo;

    var repromptText = "";

    var sessionAttributes = {};

    var shouldEndSession = false;
    
    var speechOutput = wordsToRepeat.value;

    repromptText = "";

    callback(sessionAttributes,
         buildSpeechletResponse(cardTitle, speechOutput, 
             repromptText, shouldEndSession));
}

3e. Assemble the App!

  1. Create your Lambda Function and record your ARN value.
     
  2. In Amazon's Developer Console, create a new Skill and initialize it with your ARN value, Intent Schema, Custom Slot Types (optional), and Sample Utterances.
     
  3. Assuming no errors, your Skill will validate and become testable!

4. Test your Skill

If you don't have an Echo, there's a JSON based tester in the Amazon Developers Console

https://www.youtube.com/watch?v=hJPzMQEkqIE

Resources

Made with Slides.com