Objective

To demonstrate a method to integrate Alexa Echo Dot with RDK 4.0 by developing Alexa skills and integrate on RDK 4.0 reference platforms:-

• Raspberry Pi
• Arris Accelerator - Commscope

Following sections will give the architecture details and end to end flow diagram  

Amazon Alexa Voice Control Reference High Level Architecture

Architecture ( End Product Reference Architecture for Echo Dot Integration )

*Note: Above is the stable state end goal of the product architecture. Rest of the section is for demonstration purposes as an example and not realizing the full scope of the above architecture as end goal. 

Architecture used for proof of integration concepts.

<Explain how this trimmed version helps to demonstrate echo dot integration>

RDK Echo Dot Integration - Amazon Voice Server ( AVS ) setup and configurations

Alexa Skill  Creation

Step one: got to https://developer.amazon.com/alexa/console/ and select your skill name

Step two: got to test tab

Step Three: Type or say your test command, eg: Alexa, tell stb voice move right

Step four: You will get a response, please note the user id – this maps our account to Django server later

RDK Alexa Voice Echo Dot Integration - Alexa Server configuration details

Configure Alexa User Account and STB account

Step one: Go to https://rdkvoice.iptn.io/admin

Step two: Go to  first users link 

Step Three: create a user account  

Step Four: user creation

Step Five: STB account creation

Step Five: STB account creation

E2E Sequence Diagram

API definition

method: "RDK.RegisterNewDevice"

Input:

• "id": "1234567890"

Output:

• "id": "1234567890"

method: "RDK.CompleteRegistration"

Input:

• "id": "1234567890"

Output:

• "id": "1234567890"

method: "RDK.finalizeRegistration"

Input:

• "id": "1234567890"
• availableApplications:[{}]

Output:

• "id": "1234567890"
• result: {available applications in the box}

method: "RDK.LaunchApplication"

Input:

• "id": "1234567890"
• appURI: string
• contentID: string
• searchString: string

Output:

• result : Boolean

Errors:

• error code
  • 1 Application not available
  • 2 content not available

method: "RDK.GetCurrentApplication"

Input:

• None

Output:

• The appURI of the current application

method: "RDK. Navigate"

Input:

• direction: String

Output:

• Result: success

method: "RDK. MediaControl"

Input:

• action: String

Output:

• Result: success

How to configure server

Follow the process outlined in the attached document to configure the server.

Deployment guide.docx

FAQ

8.1 Is this a ready to deploy service integration ?

No, this is not. It is only meant for demonstration of capability purposes.

8.1 What is pending from this proof of concept to actually integrate echo dot as part of RDK 4.0 Voice ecosystem ?

Scalability and security of product deployment is missing in this implementation. 

8.3 Do I have to sign any special license or pay a fee to use this service if I have Alexa Echo Dot configured in my home Wifi ?

No. 

8.4 Who shall we contact in case we have to integrate more skills to this mode ?

<<>>

8.5 Which version of Alexa devices are supported ? What about STB models ? RDK Software versions ?

<<Write the versions >>

8.6 What happens if my echo device does not recognize the commands in the prescribed formats of the API ? Which are the areas I should look for triaging ?

<Answer this by simple steps check connection, syntax, other dot commands, server response in dev console as well as logs >>

8.7 How scalable is the current solution ?

End state architecture has scalability built into it, current PoC is only for demonstration purposes.

8.8 How secure is the current solution ?

End state architecture has security built into it, current PoC is only for demonstration purposes.

8.9 Is there a plan in RDK community to realize the end state architecture ?

Currently No. We will revisit the plan based on the users feedback and interest in using this service.

8.10 Who are the key contacts for this initiative ?

<HJ's details>