如何在dialogflow中打断音乐

Question

I am building a chatbot using dialogflow. I am using music between the questions. I want the user to interrupt the music and then provide the answer to the question and then the next question to be asked.

This is the code where I am using music:

app.intent('First', (conv) => {
    const ans = conv.parameters.any;
    if(ans == 65){
            senddata[0] = qstion[0] + ans;
            const speech = '<speak><audio src="https://actions.google.com/sounds/v1/science_fiction/alien_beam.ogg"/>Did not get the audio file</speak>';
            conv.add(speech);
            conv.ask(qstion[1]);
        
    }
    else{
            conv.add('Please enter a valid number');
            conv.ask(qstion[0]);
    }
});

This is the next intent:

app.intent('Second', (conv) => {
    const ans = conv.parameters.any;
    if(ans == 108){
            senddata[1] = qstion[1] + ans;
            conv.ask(qstion[1]);
        
    }
    else{
            conv.add('Please enter a valid number');
            conv.ask(qstion[0]);
    }
});

Following is the code where I am declaring qstion and senddata:

global.sample = ['Question1, What is 33+32',
'Okay, Question2, What is 76+32',
'Okay, Last Question, Did you like the game']
global.senddata = [];

The issue here I am facing is if during the music if the user speaks "Okay Google" to interrupt the music, the music stops but the flow goes into the else part and says "Please enter a valid number". After that it asks the first question again but the flow goes to "Second" intent.

I want the flow to stay in the "First" intent itself when the user interrupts the music. Is there anyway I can do this?

Answer 1

Not the way you have designed it, but that's partly because the request/response model for Dialogflow (and Actions on Google) don't quite work that way.

For starters, it doesn't quite make sense to say you want to "stay in the Intent". Intents aren't something you are "in" - they represent a specific action from the user. For Actions on Google - this is usually representing the specific action of saying something.

So when the user "breaks in" to a response, no matter if they break in while a sound file is playing or if the speech is playing, then this is treated as the end of the response and the user generating their next request. The request from the user triggers an Intent, the Intent determines how it is handled, and the webhook sends something back.

This cycle (user input, processing based on Intent, send reply back) can't really be changed.

For playing an audio file, however, there are a couple of tricks you can pull. You can, for example, use a Media response . With this,

• At some point, you'll send back a spoken message and Media that will play after this spoken message.
• The user can break in at any time, and it will be handled through normal Intent processing
• If the user does not break in, when the media finishes playing, it will generate a actions_intent_MEDIA_STATUS event in Dialogflow. You can setup an Intent that will handle this Event and reply accordingly.