Media Response in Custom Payload (Dialogflow) - actions-on-google

Question

I'm trying to code a Media Response in a Custom Payload with no luck. I'm surely doing something wrong but I have no idea :) The Media Response does not show up when testing. (Please note that I'm trying this in an english action). Here's the JSON code:
{
"platform": "google",
"type": "custom_payload",
"payload": {
"google": {
"richResponse": {
"items": [
{
"mediaResponse": {
"mediaType": "AUDIO",
"mediaObjects": [
{
"name": "Exercises",
"description": "ex",
"largeImage": {
"url": "https://firebasestorage.googleapis.com/...",
"accessibilityText": "image..."
},
"contentUrl": "https://firebasestorage.googleapis.com/..."
}
]
}
}
]
}
}
}
}
UPDATE:
I've updated the JSON to something like this. But I get an error :"API Version 2: Failed to parse JSON response string with 'INVALID_ARGUMENT' error: ": Cannot find field."
{
"platform":"google",
"type":"custom_payload",
"payload":{
"google":{
"richResponse":{
"items":[
{
"simpleResponse":{
"textToSpeech":"Hey! Good to see you."
}
},
{
"mediaResponse":{
"mediaType":"AUDIO",
"mediaObjects":[
{
"name":"Exercises",
"description":"ex",
"largeImage":{
"url":"https://firebasestorage...",
"accessibilityText":"..."
},
"contentUrl":"https://firebasestorage.googleapis.com/..."
}
]
}
}
],
"suggestions":[
{
"title":"chips"
}
]
}
}
And here's the debug information:
{
"audioResponse": "//NExAARWG...",
"conversationToken": "GidzaW11bG...",
"debugInfo": {
"agentToAssistantDebug": {
"agentToAssistantJson": "{\"message\":\"Failed to parse Dialogflow response into AppResponse, exception thrown with message: Empty speech response\",\"apiResponse\":{\"id\":\"cd7204ac-ab80-42aa-9755-6123cbb938a6\",\"timestamp\":\"2018-03-11T09:02:35.827Z\",\"lang\":\"en-us\",\"result\":{},\"status\":{\"code\":200,\"errorType\":\"success\"},\"sessionId\":\"1520758955600\"}}"
},
"assistantToAgentDebug": {
"assistantToAgentJson": "{\"user\":{\"userId\":\"AA9douaa4XGkqtmcU_EDjPy7PQ_9\",\"locale\":\"en-US\",\"lastSeen\":\"2018-03-11T09:02:09Z\"},\"conversation\":{\"conversationId\":\"1520758955600\",\"type\":\"NEW\"},\"inputs\":[{\"intent\":\"actions.intent.MAIN\",\"rawInputs\":[{\"inputType\":\"VOICE\",\"query\":\"Talk to Zen Coach\"}]}],\"surface\":{\"capabilities\":[{\"name\":\"actions.capability.SCREEN_OUTPUT\"},{\"name\":\"actions.capability.MEDIA_RESPONSE_AUDIO\"},{\"name\":\"actions.capability.WEB_BROWSER\"},{\"name\":\"actions.capability.AUDIO_OUTPUT\"}]},\"isInSandbox\":true,\"availableSurfaces\":[{\"capabilities\":[{\"name\":\"actions.capability.SCREEN_OUTPUT\"},{\"name\":\"actions.capability.AUDIO_OUTPUT\"}]}]}",
"curlCommand": "curl -v 'https://api.api.ai/api/integrations/google?token=0def1bb6be4b4bf2810ec68bf6f37a6a' -H 'Content-Type: application/json;charset=UTF-8' -H 'Google-Actions-API-Version: 2' -H 'Authorization: eyJhbGciOiJSUzI1NiIsImtpZCI6ImFjMmI2M2ZhZWZjZjgzNjJmNGM1MjhlN2M3ODQzMzg3OTM4NzAxNmIifQ.eyJhdWQiOiJ6ZW4tY29hY2giLCJhenAiOiI0OTYwOTIwOTE1NzEtMGNhY3VtczVkZ3F1OWpkM2k0dHZpOGFiOTVydXQ2NnQuYXBwcy5nb29nbGV1c2VyY29udGVudC5jb20iLCJleHAiOjE1MjA3NTkwNzUsImlzcyI6Imh0dHBzOi8vYWNjb3VudHMuZ29vZ2xlLmNvbSIsImp0aSI6IjY4NDc0NThhNTNhZGExODAxZjMwMjAyYjkxZGIyODZhMjk1NzA2YmIiLCJpYXQiOjE1MjA3NTg5NTUsIm5iZiI6MTUyMDc1ODY1NX0.e1cqg96F5L-BvD0yJz3UFgsnX_0TRox0Lu8R9K5NhhXcQVfC7mq1QwCqs2DGrUJGquGdW2GhzBU2lzf4ro2TUeieg4ozak1OmiYAMqtiCH0EodeHy59AXXqzb3a35YuD7CmSDu6qVQRfEp8uaaH2t-Sq9lUchudNOgjucip3ex9Rr2XacHm0qWtV69H1o-Yq5INl5HHR0kNqtEIsxUox961imKvDLN5s--F35yTbAhIWibr6OmaACyzSQW5X7OjrJ2781DSmEdYn73poDbuwMS9E2l9B-QTUHAIpUM5b4WqrFkD6XKALdf2pQFwZlRRhDzRiDKWLA-i1w-mcak0LWw' -A 'Mozilla/5.0 (compatible; Google-Cloud-Functions/2.1; +http://www.google.com/bot.html)' -X POST -d '{\"user\":{\"userId\":\"AA9douaa4XGkqtmcU_EDjPy7PQ_9\",\"locale\":\"en-US\",\"lastSeen\":\"2018-03-11T09:02:09Z\"},\"conversation\":{\"conversationId\":\"1520758955600\",\"type\":\"NEW\"},\"inputs\":[{\"intent\":\"actions.intent.MAIN\",\"rawInputs\":[{\"inputType\":\"VOICE\",\"query\":\"Talk to Zen Coach\"}]}],\"surface\":{\"capabilities\":[{\"name\":\"actions.capability.SCREEN_OUTPUT\"},{\"name\":\"actions.capability.MEDIA_RESPONSE_AUDIO\"},{\"name\":\"actions.capability.WEB_BROWSER\"},{\"name\":\"actions.capability.AUDIO_OUTPUT\"}]},\"isInSandbox\":true,\"availableSurfaces\":[{\"capabilities\":[{\"name\":\"actions.capability.SCREEN_OUTPUT\"},{\"name\":\"actions.capability.AUDIO_OUTPUT\"}]}]}'"
},
"sharedDebugInfo": [
{
"name": "ResponseValidation",
"subDebugEntry": [
{
"debugInfo": "API Version 2: Failed to parse JSON response string with 'INVALID_ARGUMENT' error: \": Cannot find field.\".",
"name": "UnparseableJsonResponse"
}
]
}
]
},
"response": "Zen coach isn't responding right now. Try again soon.",
"visualResponse": {
"visualElements": []
}
}

Answer 1

Are you adding "platform":"google" and "type":"custom_payload" in the custom payload? If so, try removing that.
I made the following work with my Voice Metronome application:
{
"google":{
"richResponse":{
"items":[
{
"simpleResponse":{
"textToSpeech":"Hey! Good to see you."
}
},
{
"mediaResponse":{
"mediaType":"AUDIO",
"mediaObjects":[
{
"name":"Exercises",
"description":"ex",
"largeImage":{
"url":"http://res.freestockphotos.biz/pictures/17/17903-balloons-pv.jpg",
"accessibilityText":"..."
},
"contentUrl":"https://freepd.com/Chill/Chill Air.mp3"
}
]
}
}
],
"suggestions":[
{
"title":"chips"
}
]
}
}
}

Answer 2

The issue is that the richResponse property still needs to follow the rules of the RichResponse object. The first item in it must be a SimpleResponse object. (I haven't tested, but you can probably have that say nothing, but it is a good spot to have an introduction to your audio.)
The error message Failed to parse Dialogflow response into AppResponse, exception thrown with message: Empty speech response suggests that it might also be looking for a speech parameter on the top-level object in the response, which is what Dialogflow v1 expects to duplicate either the simpleResponse ssml or textToSpeech parameters. I'm not sure why that would appear if you're set to v2, but it sounds like something might be confused there. I would make sure you're using v1 and that you have a speech parameter.
Also keep in mind that the reviewers will look for suggestion chips about how to move the conversation forward during or after the audio if this isn't a final response.