Course info
Mar 13, 2018
1h 57m

The creation of an interactive and intelligent application is helped by the use of automatic speech recognition (ASR) and natural language understanding (NLU). In Getting Started with Amazon Lex, you will be introduced to the skills required to leverage Amazon Lex to build these user interactions for the applications you create. First, you will learn how to define the intents that the ChatBot will fulfill for the user. Next, you will explore how to create the slots that will gather the required information from your user. Finally, you will discover how to implement Lambda functions that process the information gathered by your ChatBot to complete the desire of your users. When you are finished with this course, you will be prepared to create simple informational bots using Amazon Lex.

About the author
About the author

Mike is a developer, architect and trainer and has worked with many different tools and technologies for over 20 years. When not working on, learning or sharing something to do with technology he enjoys spending time with his family, especially camping and traveling.

More from the author
Architecting for Performance Efficiency on AWS
3h 18m
Mar 1, 2019
Introduction to UML
2h 13m
Sep 9, 2013
Section Introduction Transcripts
Section Introduction Transcripts

Course Overview
Hi everyone. My name is Mike Erickson, and welcome to my course, Getting Started with Amazon Lex. I am a consultant supporting clients with both development and architectural needs. With the many digital assistants that are now available, voice interfaces are quickly becoming more mainstream. Many users find these assistants and chat bots to be very helpful in completing everyday tasks. In this course we are going to build a simple chat bot using the managed services provided by AWS with Amazon Lex. Some of the major topics that we will cover include, defining a conversational interface to our bot, using AWS Lambda to implement the functionality our bot exposes, understanding how to create new versions of our bot, allowing for continued development, and gracefully handling difficulties in understanding our users. By the end of this course you will be prepared to create simple informational bots using Amazon Lex. From here you may wish to continue preparing yourself for deeper interactions with Amazon Lex by reviewing courses on Alexa and AWS Lambda. I hope you'll join me on this journey to learn bot development with the Getting Started with Amazon Lex course at Pluralsight.

Introducing Lex, Amazon Lex
Hi. This is Mike with Pluralsight. Welcome to the course, Getting Started with Amazon Lex. In this first module we're going to introduce Lex, and that is Amazon Lex. So we'll start by getting an understanding of what Amazon Lex is, why we care, what we may want to do with it. From there we'll move on to talk about how the Echo devices and the Alexa voice services are utilized as part of Amazon Lex, and what the relationship is there. Then we'll talk about ASR and NLU. You'll understand that ASR is Automatic Speech Recognition, and that NLU is Natural Language Understanding, and we'll see where those are important for our Lex applications. Then we'll do a demo. The demo that we'll do is a prepackaged solution from AWS that shows us some of the features of Amazon Lex. Finally, we'll come back to the slides and we'll introduce the sample that we'll be creating throughout the rest of the course, so let's jump in and get going with understanding what is Amazon Lex?

Building Your First Amazon Lex Chatbot
Hi. This is Mike Erickson with Pluralsight back again for the next module in Getting Started with Amazon Lex. In this module we're going to go ahead and start building our own chatbot. To be able to build our own chatbot we have to understand a few topics. We saw these in the demo that we did in the first module. The first thing we need to understand is what is an intent, which lets us identify what we want the bot to do for us. Then we have to understand what utterances are, which allow the bot to understand what our intent is. Then we'll start to understand slots. Slots allow us to provide parameters to our bots, so that it can solve our requests, and finally, we'll start to understand prompts, which is how the bot asks us for the information to fill in the slots.

Improving Your Prompts
Hi. This is Mike Erickson back for the next module of Getting Started with Amazon Lex. In this module we're going to see what we can do to improve how the bot interacts with our users. We'll look at the prompts and the utterances that we've been working with and see how we can make them better. We'll start this module looking at our utterances. We'll see how we can enhance those, and how we can add slot values to the utterances. Then we'll look at the slots themselves, and see how we can make them better, how we can improve the prompts that we provide to the users, and other things we can configure within the slot.

Adding Fulfillment
Hi. This is Mike with Pluralsight. Welcome to the next module of Getting Started with Amazon Lex, and in this module we're going to see how we can add fulfillment to our bot. The first thing we will look at is the message formats. There are two message formats that we need to understand. The first is the format in which the bot sends the information to our lambda function, and the second is the format in which we must return our response to the bot. Then we're going to go ahead and build our own lambda function that will fulfill our FindDoctor intent, and return a suggested doctor to our user.

Managing Your Chatbot Lifecycle
In this module we're going to look at how we can manage our chatbot lifecycle to understand how we can continue developing our chatbot even after we've deployed it. To do this we'll start by looking at how we can make different versions of our bot and create aliases that reference those versions, and we'll see why that's important. Then we'll go ahead and do a demo where we publish our chatbot and see how it interacts with versions and aliases.

Adding Error Handling
In this module we're going to look at how we can interact with users that we have a hard time understanding. First, we'll learn what error messages and hang-up phrases are, and how they can be used to have a more friendly interaction with our users. Then we'll return to the Amazon Lex console, and we will add these error handling procedures to our bot.