Alexa Skills Kit > Dialog Management
Alexa Conversations is a new AI-driven approach to dialog management that enables you to create skills that customers can interact with in a natural, less constrained way - using the phrases they prefer, in the order they prefer – while freeing you to focus on the highest value parts of your experience.
Conversations are more nuanced than simply understanding words and sentences. The number of ways customers can engage and the variety of dialog paths they can take results in a combinatorial explosion of states and code. For example, a pizza skill with seven topping combinations might require more than 5,000 dialog paths just to place an order!
Alexa Conversations manages the dialog to collect inputs so you can focus on tasks.
Alexa Conversations uses deep learning to extrapolate the dialog paths that customers might successfully follow through a vast range of possible conversations. Then it tracks context, inputs, and state in real-time to automatically complete an interaction. Alexa can manage a dialog that collects information to help your skill call external APIs, leaving you to focus on designing the overall experience. Now you can stop managing flowcharts and start using AI training data to expand what your skill can do.
The Art Museum skill combines Alexa Conversations dialog management with the Art Institute of Chicago's public API. Notice how this interaction can be long, open-ended, and doesn't require the user to follow strict prompts or guidance. Users are able to select topics on their own and change their minds, and the the experience feel more natural because Alexa manages the dialog on the fly. Learn more.
"Alexa Just Keeps Up"
"Game-Changer for Handling Inputs"
Alexa Conversations keeps track of state, maintains context, accepts customer-driven corrections, and then calls your services at the right time. Unlike traditional dialog managers, with Alexa Conversations you do not have to code or even identify all of the different ways customers might interact. Instead, you provide a set of example dialogs and how they map to specific services in your skill.
This is how a weather skill using Alexa Conversations performs a task to get data about Seattle by invoking an API.
const GetWeatherApiHandler = {
canHandle(handlerInput) {
return util.isApiRequest(handlerInput, 'GetWeather');
},
async handle(handlerInput)
//make an API call or look something up in a database
//relay the data from your skill code to your front-end API definition in the format you defined
return {
apiResponse: {
high: '70',
low: '45',
city: 'Seattle'
}
};
};| Natural Interaction | Alexa Conversations Supports |
|---|---|
| "Medium two-topping” | Accepting multiple values at once |
| "How many people can that feed?” | Context carryover and anaphora |
| "Pepperoni and onion” | Selecting multiple values from a list |
| "No, make that a large” |
Self-correction |
| “Apply the third coupon” | Selecting from a visual or vocalized list |
| Tapping choices on a screen |
Multimodal interaction |
Skills that feel natural work with a wide range of phrases and unexpected dialog turns and have the memory to sustain long back- and-forth sessions. Your skills need to flex while gathering inputs and accept too much or too little information out of sequence. Skills need to automatically track across topical shifts, reference context, and adjust to corrections.
Alexa Conversations can give your skills the capacity for natural conversational interactions.
