Step 8: Test for Certification (VSK Echo Show)
Integrate the VSK into a Multimodal Device
The following guidelines detail the user experience requirements to certify your skill against. Review these guidelines carefully to ensure that your multimodal device video skill meets the certification requirements. The guidelines here include a Certification Checklist of the individual test cases that Amazon reviews your skill against for skill launch quality.
- Prerequisites and Notes
- Targeting Your Video Skill on Multimodal Devices
- Overall Certification Requirements
- Testing Utterance Support
- Key Terms
Prerequisites and Notes
As part of the capability discovery payload, you declare the capabilities of your skill, such as channel change, quick play, and search. Your skill then needs to pass certification for those capabilities, following the guidelines listed. As you are creating your skill, list the capabilities that you plan to support in the testing instructions field.
This certification checklist provides a full list of supported utterances. Make sure you test your skill with all of the supported utterances before you consider your integration to be complete. Additionally, to support "play by title" and "search by title" utterances for video-on-demand content, you must complete Catalog Integration with Amazon. If you have not already integrated your catalog with Amazon, contact your Amazon Fire TV Solution Architect, and have them work on the integration process.
Targeting Your Video Skill on Multimodal Devices
When you test your skill on a multimodal device, make sure you target your skill through any of the following methods:
Explicitly target your skill: You can explicitly target your skill in your utterances by saying, "Alexa, play X on <video provider>" or "Alexa, show me comedies on <video provider>."
Launch your skill from Video Home: You can find your skill in Video Home by saying either "Alexa, show me videos" or "Alexa, go to video home," and then tapping your skill icon.
Launch your skill by means of voice: You can also open your skill by voice by saying "Alexa, open <video provider>."
When you target your skill like this, Alexa performs commands and other lookups within the context of your app and catalog, rather than looking across all video skills on the device.
Overall Certification Requirements
Your skill must meet the requirements listed in the following sections to pass certification.
Latency Requirements: Your skill's end-to-end latency should be comparable to or better than currently-live video skills on multimodal devices.
Web Player Requirements: Your web player must meet the following requirements to pass certification.
Closed Captions: Multimodal devices devices have a setting in which a user can toggle closed captions on and off across all video experiences. To create a unified experience across all players, video providers must adhere to this setting. If the user toggles the device setting On, the player should also toggle closed captions to appear within the video session. If the user toggles closed captions On in the player, the player needs to broadcast that so that the user can toggle the device setting. For more information about how to broadcast closed captions capabilities and touch events, see Video Skill API Reference.
Continuous Play for TV Series VOD: If a user is currently watching a TV series, continuous play is recommended and supported. Once a new episode is playing, you must broadcast the capabilities of that next episode. If it is a single episode, you must honor the user's request and close the session.
Testing Utterance Support
Your video skill must provide appropriate responses to the utterances you support. To determine what capabilities your skill provides, Alexa sends the
Discover directive to your Lambda. You list your supported capabilities in the
capabilities section in the
Discovery Interface for more details.
If your app doesn't support a capability (for example, you might not support
ChangeChannel if your app doesn't have channels), you don't have to certify against those utterances. And if you don't claim support for
ChangeChannel directives, Alexa won't send those directives to your Lambda.
However, if your skill only supports certain capabilities for an interface, your responses should provide the appropriate handling for unsupported actions.
See the Utterances Reference for a detailed list of the various utterances that Echo Show devices support, including each phrase by locale as well. Utterances are grouped by the requests that Alexa sends when it hears the utterance. These groups are as follows:
- GetPlayableItems Directive Utterances
- GetDisplayableItems Directives Utterances
- Channel Navigation Directives Utterances
- Transport Control Directives Utterances
The following terms are used in the certification checklist below:
- Voice command that does not specify the content provider or app name, e.g. "Watch Bosch".
- Voice command that specifies the content provider or app name, e.g. "Watch Bosch on Amazon Prime".
- quick play
- Play content directly by title, actor, genre, sports, teams, leagues without stopping at a detail page.
- quick play – implicit
- Quick play phrase that does not mention the content provider or app name in the play phrase.
- quick play – explicit
- Quick play phrase that mentions the content provider or app name in the play phrase.
- voice cliff
- Voice experience that leads to the user having to pick up the remote control.