Announcing Test Workbench for Amazon Lex

Amazon Lex is a 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 Test workbench on Amazon Lex that enables customers to author and execute test sets to measure bot performance and expedite bot development.

Customers need to continuously update their bots as they add new use-cases or improve the conversational experience on existing use-cases. With each update, customers need to ensure that their bot meets the performance criteria. Until now, customers had to manually create test sets by curating conversational data and write custom scripts to execute tests. These tests generated a limited set of non-standardized metrics that didn’t provide sufficient information to debug and improve the bot. In addition, they had to build and manage the infrastructure for the test suite. With Test workbench, customers can create and execute test sets with just a few clicks, expediting bot development and testing. It enables customers to automatically generate audio and text-based test sets from user interactions, saving time and manual effort. Customers can also upload custom curated test conversations in CSV format. The results from test execution provide aggregated and detailed insights on speech transcription, intent matching, and slot resolution.

Test workbench is generally available in all the AWS regions where Amazon Lex V2 operates. To learn more, read our blog, view the explainer video, or visit the documentation page.