Understand How Users Interact with Skills

When a user speaks to a device with Alexa, the speech is streamed to the Alexa service in the cloud. Alexa recognizes the speech, determines what the user wants, and then sends a structured request to the particular skill that can fulfill the user's request. All speech recognition and conversion is handled by Alexa in the cloud.

Every Alexa skill has an interaction model defining the words and phrases users can say to make the skill do what they want. This model determines how Alexa communicates with your users.

The following sections provide examples and more detail around how users communicate with Alexa and what you need to do as a developer when designing a skill.

Note that the interaction model described here is focused on voice interaction. If you are developing a skill that includes screen display, video, or touch interaction on an Alexa-enabled device with a screen, also refer to Best Practices for Designing Skills for Alexa-Enabled Devices With a Screen. For details about adhering to Alexa marketing and branding guidelines with your skill, see Amazon Alexa Brand External Guidelines.

How Users Interact with Alexa
What is an Interaction Model?
Examples of Interaction Models
Next: Learn what you need to get started
Jump In

How Users Interact with Alexa

End users interact with Alexa's abilities in the same way – by waking the device with the wake word (or a button for a device such as the Amazon Tap) and asking a question or making a request.

In addition, with Alexa-enabled devices with a screen, the end user can touch the screen to interact with Alexa, if the skill supports this interaction.

For example, users interact with the built-in Weather service like this:

User: Alexa, what's the weather?
Alexa: Right now in Seattle, there are cloudy skies…

Interacting with a custom-built skill to look up tide information is very similar, although the user can include a name identifying the skill ("tide pooler" in this example):

User: Alexa, get high tide for Seattle from Tide Pooler
Tide Pooler: Wednesday February 17th in Seattle, the first high tide will be around 1:42 in the morning, and will peak at about 10 feet…

Similarly, a user can tell a skill to control a particular cloud-enabled device such as a light. In this case, the light must be connected to the Internet (such as via a smart home hub). The user names the specific light ("living room lights" in this example):

User: Alexa, turn on the living room lights
A light previously configured and named living room lights is turned on.
Alexa: OK.

What is an Interaction Model?

In the context of Alexa, an interaction model is somewhat analogous to a graphical user interface in a traditional app. Instead of clicking buttons and selecting options from dialog boxes, users make their requests and respond to questions by voice:

Action	Voice User Interface (Interaction Model)	Typical Graphical User Interface
Make a request	User says, "Alexa, get high tide from tide pooler."	User clicks a button.
Collect more information from the user	Alexa replies, "For what city?" and then waits for a response.	App displays a dialog box and waits for user to select an option.
Provide needed information	User replies, "Seattle."	User selects options and chooses OK.
User's request is completed	Alexa speaks the requested information: "Wednesday February 17th in Seattle, the first high tide will be…""	App displays the results of the request.

When users speak questions and make requests, Alexa uses the interaction model to interpret and translate the words into a specific request that can be handled by a particular skill. The request is then sent to the skill.

You define your own interaction model when creating a custom skill. The Smart Home Skill API, Video Skill API, Music Skill API, and others provide a built-in interaction model.