Building a Language Translation Bot

With Amazon Lex

Objective

The objective of this project is to develop a language translation bot. With so many translation tools already available to the public, you might wonder—why create another one?The key advantage of a custom translation bot lies in its ability to serve specific business needs. Unlike generic translation services, a personalized bot can be tailored to reflect a company’s unique requirements, industry-specific terminology, and preferred languages. Additionally, it enhances data security by keeping sensitive information within a controlled environment, reducing reliance on third-party services that may store or process data externally.

Project Sequence

Creating an empty chatbot
Specifying intents and slots
Specify Fulfilment
Create an IAM role
Create a Lambda function
Test the Lambda function
Test the chatbot

Step 1: Create the Chatbot in Amazon Lex

In the Amazon Lex console, select “Create bot”.
Choose Traditional / Create a blank bot
‍

Name the bot and select the first option for IAM permissions. Choose "No" for COPPA.
Set English as the language and keep default settings. This completes the creation of an empty chatbot.
‍

Step 2: Configure the Bot for Translation

Image 1

Image 2

Image 3

Define the bot’s Intent, which represents its function. Since this bot translates text, name the intent "translatorintent"
Add an Utterance, which acts as a key phrase that triggers the chatbot.
Set up Slots to process translation requests:
‍
Create a slot type outside the Intent console, listing the languages the bot will translate to (see image 1).

Inside the Intent editor, add the slot type and a prompt asking for the target language (see image 2).

Add a second slot to capture the text input for translation (see image 3).
Optionally, set an initial response
Define fulfillment outputs for success or failure
Enable the Lambda function for processing
Add a closing response to finalize the intent

Step 3: Create an IAM Role for Lambda Execution

In the IAM console, create a new role for the Lambda service

Attach the following policies: TranslateFullAccess, AWSLambdaBasicExecutionRole
Name, review, and create the role.

Step 4: Create and Configure the Lambda Function

Image 4

Image 5

Image 6

In the AWS Lambda console, create a new function and attach the IAM role created earlier.
Set the runtime to Python 3.13 (see image 4).
Edit the Lambda function code to process translations (see image 5).
Deploy and test the Lambda function:
Configure a Test Event (see image 6).

Run the test to ensure it succeeds.

Step 5: Integrate and Test the Chatbot

In the Amazon Lex console, build the Intent

Under Testing settings, add the Lambda function created earlier.
Test the chatbot by interacting with it.