This project is a serverless multi-language translation chatbot built on AWS.

🎥 Demo

Here’s a quick look at my AWS Translation Bot in action:

👉 Watch the full demo on YouTube

Architecture

User → Amazon Lex V2 → AWS Lambda (Python) → Amazon Translate → Response back to Lex → User ↘︎ CloudWatch Logs (Monitoring) The solution leverages a fully serverless design:

1. Amazon Lex V2 – Captures user input, interprets intents, and triggers fulfillment.

2. AWS Lambda (Python 3.12) – Acts as the fulfillment code hook. Handles slot data, invokes Amazon Translate, and returns translated text back to Lex.

3. Amazon Translate – Provides language translation service.

4. Amazon CloudWatch – Logs events, errors, and metrics for debugging and monitoring.

5. IAM Roles & Policies – Securely allow Lex to invoke Lambda and Lambda to call Amazon Translate.

6. Features

• Natural Language Understanding with Lex intents and slots.
• Custom Lambda Fulfillment written in Python.
• Multi-language Translation using Amazon Translate API.
• Cloud-native, serverless and scalable design.
• Error Handling & Logging integrated with CloudWatch.

Tech Stack

• AWS Services: Amazon Lex V2, AWS Lambda, Amazon Translate, IAM, CloudWatch
• Language: Python 3.12
• Infrastructure: Serverless (fully managed AWS services)

Setup

1. Create Lambda Function

   • Runtime: Python 3.12
   • Add permissions: TranslateText, CloudWatch Logs, and allow Lex to invoke.

2. Deploy Lambda Code
   Upload the lambda_function.py (see code in repo).

3. Configure Amazon Lex

   • Create a bot with an intent (TranslateText).
   • Add slots: Text, LanguageCode.
   • Enable fulfillment with the Lambda function.

4. Test & Debug
   Use the Lex test console. Monitor errors via CloudWatch Logs.