Amazon Lex is aENnbsp;service for building conversational interfaces into any application using voice and text. With Amazon Lex, you can quickly and easily build conversational bots (“chatbots”), virtual agents, and interactive voice response (IVR) systems. Today, we are excited to introduce utterance generation for Amazon Lex that expedites bot development by leveraging generative AI to automatically produce new training utterances.

Amazon Lex’s machine learning model uses utterances to recognize and respond to the user intent. Bot developers can add sample training utterances to their bots to improve intent classification, including user input called slots (“what is the balance in my {AccountType}”). Traditionally, bot developers defined a representative set of utterance samples and manually associated the utterances with an intent. With utterance generation, customers can leverage foundational models on Amazon Bedrock and generateENnbsp;training utterances with just one click. This feature gives bot developers and conversation designers the ability to edit or delete the generated utterances before accepting them. The feature works with both new and existing intents.

Utterance generation is now generally available in the English language in US East (N. Virginia) and US West (Oregon) regions. Amazon Connect and Amazon Bedrock customers can use this feature in any Amazon Lex bot. To learn more, visit the documentation page.