If you are using Amazon Lex to train your virtual assistant, you must have realized how easy it is. In a short period of time, you can have a virtual assistant based on Artificial Intelligence answering your clients’ questions.
Configuring questions and answers in Amazon Lex is very easy, but deploying your chatbot to your website can be really complicated.
To be able to do that, you need to have advanced knowledge in programming to make the connection with the API, to know how to deal with errors and to use suitable methods.
Also, you need to have skills to design a chat box, from its functionality to the UX, the integration with the API and the installation in the website.
With Blahbox, you can solve these issues with just a few clicks.
You can watch the following video to see what this is about:
You can find below my explanation, step by step, about how to create the first chat box in Blahbox:
First, you must have an account in Blahbox.
1. Go to blahbox.net
2. Then, click on the “Start for free” button.
3. Register using your Google or Facebook user. You can also register with an email if you wish to.
Create your first chat box
1. When you go into Blahbox for the first time, you have to click on the “New chat” button.
2. Then, you need to choose the option “For a WebSite”.
3. Write the name you want to give to your chat box and then click on “Go”.
4. Here, I recommend you to click on the button “Skip step, set up later”. I will explain later how to insert the credentials for Amazon Lex.
And that’s it! In that moment, you’d have your first chat box created. You can access the chat box from Blahbox’s dashboard.
Now, we’ll connect the chatbot to Blahbox. Below, I’ll explain the process step by step:
Basically, what we have to do is to find the following information in Amazon Lex to paste it in Blahbox:
1. User’s Credentials
2. Bot’s name, Alias and Version
Link Amazon Lex with Blahbox
In order to link your Bot trained in Amazon Lex, you must enter into the NLP section of Blahbox’s dashboard:
1- Log into Blahbox.
2- Enter into the chat you wish to link with Amazon Lex.
3- Then, click on the option “NLP” from the side menu.
4- Select “Amazon Lex” from the options on the right side of the screen.
Amazon Lex Credentials
Once you get to this screen, you must follow the instructions below to obtain the information from AWS:
How can I obtain the User’s Credentials?
In order to obtain the user’s credentials, a new user must be created in AWS. I’ll explain to you how to do this.
Trending Bot Articles:
1. Create a group
1.1- Go into the AWS Management Console.
1.2- Go into the IAM service (inside the “Security, Identity and Compliance” group).
1.3- Go to “Groups” from the side menu.
1.4- Click on the “Create New Group” button.
1.5- Give a name to the group (e.g. LexBlahbox) and then, “Next Step”.
1.6- Find the “AmazonLexRunBotsOnly” policy.
1.7- Tick the policy as is shown in the image above and click on the “Next Step” button.
1.8- Click on “Create Group”.
2. Create a User
2.1- Go to the “Users” option from the side menu.
2.2- Click on the “Add user” button.
2.3- Give a name to the user (e.g. Blahbox).
2.4- Tick the option “Programmatic access”.
2.5- Click on the “Next: Permissions” button.
2.6- Tick on the group created in step 1 of these instructions.
2.7- Click on the “Next: Tags” button.
2.8- Click on the “Next: Review” button.
2.9- Click on the “Create user” button.
2.10- You can download the credentials from the button “Download .csv”.
2.11- Save the following information: “Access key ID” and “Secret access key”.
2.12- Once the information is saved, click on “Close”.
2.13- Paste the “Access key ID” obtained in AWS in the field “Access key ID” of Blahbox’s screen.
2.14- Paste the “Secret access key” obtained in AWS in the field “Secret access key” of Blahbox’s screen.
You’re almost there!
How can I obtain the Bot’s name, Alias and Version?
1.1- Go into the AWS Management Console.
1.2- Go to the Amazon Lex service (inside the “Machine Learning” group).
1.3- Go into the bot you wish to link with Blahbox.
1.4- Click on the “Settings” tab.
1.5- You can find the name of the bot in the top left-hand corner.
1.6- Clicking on the option “Aliases”, you’ll be able to find the bot’s aliases.
1.7- From this screen, you’ll be able to obtain the Alias and the Version of the bot you wish to link with Blahbox.
1.8- You must paste the information that you obtained, in Blahbox’s NLP screen.
1.9- Finally, you must obtain the AWS Region where the bot which was created is located.
1.10- In the top right-hand corner of the AWS Console, you can see the region where the Bot is created.
1.11- You must select the same region in Blahbox’s NLP screen.
1.12- Then, click on “Go” and that’s it! You have your chat box linked with your Amazon Lex Chatbot!!!
Now you have your first chat box created in Blahbox linked with your chatbot created with Amazon Lex! Congratulations!
Now, the last step is designing the appearance of the chat box to obtain the appearance you want for your website.
When you go to the chat configuration section in Blahbox, you can find many features to configure: fonts, texts, avatars, colors, forms, etc.
Below you can find some screenshots that will show you how easy it is to configure the appearance of your chat box.
Avatar configuration. You can use a predefined image or upload yours
Edges of the header
Format and colors of the chat bubbles
You’re almost there!
Now you have to publish the chatbot widget with the “Publish” button.
And, finally, copy and paste in your website the code that Blahbox generates automatically to deploy the chat box to your website.
You can find the code by clicking on the option “Install” from the left navigator. You have to copy the code and paste it as it is between the tags <head> or <body> of each page of your site where you want your chatbot to be present.
Install your widget
And it’s ready!
In a short period of time you have your Amazon Lex chatbot running on your website with the shapes and colors you need!
Don’t forget to give us your 👏 !
Connect Amazon Lex chatbot on your website with no code was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.