CODEWITHSUNDEEP
javaeclipseamazon-web-servicesaws-lambdaalexa-voice-service
Yvonne Marggraf
Yvonne Marggraf

Reputation: 408

How to create a Multi-Language Alexa Skill using Java?

I'm quite new to VUI and Alexa. I developed a good working Skill for Alexa Voice Services. Now I want to add a second language. I've found a tutorial in the develop documation, but I'm using Eclipse and Java to create my Lambda function like this walkthrough. The problem is, that I have no idea how to enable a second language option in the Lambda function. I have to use the same Lambda function for both languages.

My StreamRequestHandler: 

public class ApiOmatBlogSpeechletStreamRequestHandler extends SpeechletRequestStreamHandler {
private static final Set<String> supportedApplicationIds = new HashSet<String>();
static {
    /*
     * This Id can be found on https://developer.amazon.com/edw/home.html#/ "Edit" the relevant
     * Alexa Skill and put the relevant Application Ids in this Set.
     */
    supportedApplicationIds.add("amzn1.ask.skill.xxxxxxxx");
}

public ApiOmatBlogSpeechletStreamRequestHandler() {
    super(new ApiOmatBlogSkillSpeechlet(), supportedApplicationIds);
    System.out.println("Super ApiOmatBlogSpeechletStreamRequestHandler");
}

}

My Spechlet:

public SpeechletResponse onIntent(IntentRequest intentRequest, Session session) {
    Intent intent = intentRequest.getIntent();
    String intentName = (intent != null) ? intent.getName() : null;
    System.out.println("onIntent requestId={ " + intentRequest.getRequestId() + " }, sessionId={ "
            + session.getSessionId() + " } ");

    Integer step = (Integer) session.getAttribute("step");
    System.out.println("IntentName= " + intentName + " | step = " + step);

    if ("AMAZON.HelpIntent".equals(intentName)) {
        return getHelpResponse();
    } else if ("AMAZON.StopIntent".equals(intentName)) {
        return getStopResponse();
    } else if (step != null) {
        return testing(intent, session, step);
    } else {
        if ("TestIntent".equals(intentName)) {
            step = 1;
            session.setAttribute("step", step);
            return testing(intent, session, step);
        } else {
            SsmlOutputSpeech speechText = new SsmlOutputSpeech();

            speechText.setSsml("<speak> " + "The intent is invalid." + "Please repeat your demand. "
                    + "<break time='0.5s'/> " + "  </speak>");

            // Create reprompt
            PlainTextOutputSpeech speech2 = new PlainTextOutputSpeech();
            speech2.setText("I'm sorry. Please repeat your statement.");
            Reprompt reprompt = new Reprompt();
            reprompt.setOutputSpeech(speech2);

            return SpeechletResponse.newAskResponse(speechText, reprompt);
        }
    }
}

/**
 * This function will be called if you say 'start'
 */
public SpeechletResponse onLaunch(final LaunchRequest request, final Session session) throws SpeechletException {
    System.out.println(
            "onLaunch requestId={ " + request.getRequestId() + " }, sessionId={ " + session.getSessionId() + " } ");

    return getHelpResponse();
}

Thank you

Upvotes: 0

Views: 663

Answers (2)

Yvonne Marggraf
Yvonne Marggraf

Reputation: 408

It's quite difficult to settle it up. I solved it by asking, which language the user speaks.

In the Spechlet.java I added the code:

private Boolean german;
    public SpeechletResponse onIntent(IntentRequest intentRequest, Session session) {
        if (intentRequest.getLocale().getLanguage().equals(new Locale("de").getLanguage())){
            System.out.println("Spoken language == Deutsch");
            german=true;
        }else{
            System.out.println("Spoken language == Englisch");
            german=false;
        }
...
}

and follwing I can work with the Boolean german and set up my response.

If there is a better solution, which works on two languages, than I would looking forward to a posted answer or comment. But for the moment this is the best workaround to handle multiple language in Java.

Upvotes: 1

Vijayanath Viswanathan
Vijayanath Viswanathan

Reputation: 8541

You don't need different Lambda logic to support each language. You can create multiple Lambda functions based on the region to avoid latency. If you choose to deploy your code to multiple endpoints, you must maintain the same code at all endpoints. The code must handle all languages you support. The only setup you need to do is to add Alexa supported language in Alexa Skill Information, which is marked below,

enter image description here

Please read official documentation here

Upvotes: 0

Related Questions