article

KirkC@Amazon avatar image
KirkC@Amazon posted

Create Alexa Skills That Feel More Natural With Alexa Conversations

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, unconstrained way - using the phrases they prefer, in the order they prefer – while freeing you to focus on the highest value parts of your experience.

Learn more:

alexa-conversationsalexa-skills-challenge
1595522414446.png (20.8 KiB)
3 comments
10 |5000

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

This is just what I've been waiting for, but I cannot even get the sample conversation project to run without an Internet server/service error. My question is posted here https://forums.developer.amazon.com/questions/229820/new-conversations.html

0 Likes 0 ·

Very exciting.


I'm collecting health data. I don't know if the health data is "in range" until a user completes all the conversation slots and hits my API (it is different for different people, based on their history, e.g. someone's weight can't change by 20 pounds in a day).

In CONVERSATIONS, I can see how to handle a successful "Thanks for recording your {measurement} of {measurement_value}."


How do I handle my response "Your last {measurement} was {history} about {history_days} days ago. Are you sure you meant to say {measurement_value}?" and then wait for a YES or NO (or synonyms).


Thanks!



0 Likes 0 ·

Hi @LIGHTHOUSE. A similar question was asked in another thread here:
https://forums.developer.amazon.com/comments/230077/view.html

The short answer though is that you can add a "where" key to an APLA document item to conditionally choose a response based upon the API's returned values.

For future reference, when asking a new question, please click "Ask A Question" from the upper right to create a new post. Otherwise your question may not bee seen.

0 Likes 0 ·

Article

Contributors

tia@Amazon contributed to this article KirkC@Amazon contributed to this article