aswetlow
commented
3 years ago Hey @acerbisgianluca Could you provide the response JSON? How do you make APL-A response? I have a guess.
Closed acerbisgianluca closed 3 years ago
aswetlow
commented
3 years ago Hey @acerbisgianluca Could you provide the response JSON? How do you make APL-A response? I have a guess.
acerbisgianluca
commented
3 years ago Hey @aswetlow! This is the correct response. If I say "repeat" after this, I get the error reported in the first comment.
"response": {
"shouldEndSession": false,
"outputSpeech": {
"type": "SSML",
"ssml": "<speak></speak>"
},
"reprompt": {
"outputSpeech": {
"type": "SSML",
"ssml": "<speak><audio src=\"URL\"></audio> <audio src=\"URL\"></audio></speak>"
}
},
"directives": [
{
"type": "Alexa.Presentation.APLA.RenderDocument",
"token": "8",
"document": {
"version": "0.8",
"type": "APLA",
"mainTemplate": {
"parameters": [
"payload"
],
"item": {
"type": "Sequencer",
"items": [
{
"type": "Audio",
"source": "URL"
},
{
"type": "Audio",
"source": "URL"
}
]
}
}
}
},
{
"type": "Display.RenderTemplate",
"template": {
"backButton": "HIDDEN",
"type": "BodyTemplate7",
"image": {
"contentDescription": "Launch",
"sources": [
{
"url": "URL",
"size": "X_SMALL"
},
{
"url": "URL",
"size": "SMALL"
},
{
"url": "URL",
"size": "MEDIUM"
},
{
"url": "URL",
"size": "LARGE"
},
{
"url": "URL",
"size": "X_LARGE"
}
]
},
"token": "Launch",
"backgroundImage": {
"contentDescription": "background",
"sources": [
{
"url": "URL"
}
]
}
}
}
]
}I add the directive using this.$alexaSkill.addDirective(aplaDirective.build()); where aplaDirective.build() returns the directive object as you can see it in the response.
aswetlow
commented
3 years ago I tried it with the following code and it worked:
LAUNCH() {
this.$alexaSkill.addAplDirective({
type: 'Alexa.Presentation.APLA.RenderDocument',
token: '8',
document: {
version: '0.8',
type: 'APLA',
mainTemplate: {
parameters: ['payload'],
item: {
type: 'Sequencer',
items: [
{
type: 'Audio',
source:
'<AUDIO>',
},
],
},
},
},
});
return this.ask(' ', ' ');
},RepeatIntent in my handler:
RepeatIntent() {
this.repeat();
},This is the db entry in my local db.json
{
"userId": "-",
"userData": {
"data": {},
"context": {
"prev": [
{
"response": {
"speech": " ",
"reprompt": " ",
"output": {
"Alexa": {
"Apl": [
{
"type": "Alexa.Presentation.APLA.RenderDocument",
"token": "8",
"document": {
"version": "0.8",
"type": "APLA",
"mainTemplate": {
"parameters": [
"payload"
],
"item": {
"type": "Sequencer",
"items": [
{
"type": "Audio",
"source": "<AUDIO>"
}
]
}
}
}
}
]
},
"ask": {
"speech": " ",
"reprompt": " "
}
}
},
"request": {
"timestamp": "2021-04-01T10:16:53Z",
"inputs": {},
"intent": "LAUNCH"
}
}
]
}
}
}What's in your db.json, after the first request?
acerbisgianluca
commented
3 years ago @aswetlow Thank you for your example, I think I've found the problem.
You return this.ask(' ', ' ') while I'm returning this.ask(this.$speech, this.$reprompt), but for the speech part I'm using APLA so this.$speech is empty and it's stored in dynamo as null. When I call this.repeat() it throws the error ("Cannot read property 'replace' of null").
"response": {
"output": {
"ask": {
"speech": null,
"reprompt": "<audio src=\"URL\"></audio> <audio src=\"https://URL\"></audio>"
},
"Alexa": {
"Apl": [
{
"type": "Alexa.Presentation.APLA.RenderDocument",
"token": "5",
"document": {
"type": "APLA",
"mainTemplate": {
"parameters": [
"payload"
],
"item": {
"type": "Sequencer",
"items": [
{
"type": "Audio",
"source": "URL"
},
{
"type": "Audio",
"source": "URL"
}
]
}
},
"version": "0.8"
}
}
],
"DisplayTemplate": {
"type": "Display.RenderTemplate",
"type": "BodyTemplate7",
"token": "Launch",
.........
}
}
}
},
"speech": null,
"reprompt": "<audio src=\"URL\"></audio> <audio src=\"URL\"></audio>"
}So I added this.$speech.addText(' ') only when the request is coming from Alexa and now it's working!
aswetlow
commented
3 years ago Great!
You could also use $speech.addBreak('10ms');. Sometimes it looks cleaner than white space. Same same, but different 😂
I'm submitting a...
Expected Behavior
At the end of the RepeatIntent handler, I return
this.repeat()and the voice assistant should play the last output.Current Behavior
It works on Google Assistant, but doesn't work on Alexa. Probably this is related to the fact that I use APLA for
outputSpeechon Alexa, so the field in the response is empty and Jovo doesn't take into account the APLA directive.Error log
Your Environment