Add Utterance Sets for Alexa Conversations

When you configure your Alexa Conversations skill, you create utterance sets to group different ways that your user might respond. For each utterance set, you provide a list of sample utterances.

For example, when Alexa asks the user, "Do you want pickup or delivery?" in a pizza-ordering skill, the user might respond, "I want pickup," "I want to pick it up," "How about pickup," and so on. Those utterances all represent the same action, so you provide them as sample utterances to the same utterance set.

Alexa Conversations uses the utterance sets as training data to improve the machine learning model. When you populate the sample utterances for each utterance set, provide as many variations as you can think of. User utterance variations are one of the key levers by which you can improve the general quality of the dialog experience.

Get Started

To get started with Alexa Conversations, you can step through a pet match skill-building tutorial, download the code for a pizza-ordering reference skill, or use an Alexa-hosted sample skill template that includes an Alexa Conversations skill configuration and back-end skill code.


For examples of how to create utterance sets while annotating a dialog, see Step 3: Annotate the first user turn and Step 4: Annotate the user turn that provides information in Tutorial: Annotate a Dialog for Alexa Conversations.

Was this page helpful?

Last updated: Nov 27, 2023