AWS Builder Center

Builder Center

Building a WhatsApp genAI assistant with Amazon Bedrock

Building a WhatsApp genAI assistant with Amazon Bedrock

Chat in any language with an LLM on Bedrock. Send voice notes and receive transcripts. With a small tweak to the code, send the transcript to the model.

Elizabeth Fuentes

Developer Advocate/SDE, GenAI
 Spanish version 
With this WhatsApp app, you can chat in any language with an LLM on Amazon Bedrock. Send voice notes and receive transcriptions. By making a minor change in the code, you can also send the transcription to the model.
Your data will be securely stored in your AWS account and will not be shared or used for model training. It is not recommended to share private information because the security of data with WhatsApp is not guaranteed.
✅ AWS Level: Intermediate - 200
Prerequisites:
💰 Cost to complete:

How The App Works

flow


1- Message input:

Message input

  1. WhatsApp receives the message: voice/text.
  1.  Amazon API Gateway  receives the message from the  WhatsApp webhook  (previously authenticated).
  1. Then, an  AWS Lambda Functions  named  whatsapp_in  processes the message and sends it to an  Amazon DynamoDB  table named whatsapp-metadata to store it.
  1. The DynamoDB table whtsapp-metadata has a  DynamoDB streaming  configured, which triggers the  process_stream  Lambda Function.

2 - Message processing:

Text Message:

Message processing - Text Message

 process_stream  Lambda Function sends the text of the message to the lambda function named  langchain_agent_text  (in the next step we will explore it).

Voice Message:

Message processing - Voice Message

  1. The  audio_job_transcriptor  Lambda Function is triggered. This Lambda Function downloads the WhatsApp audio from the link in the message in an  Amazon S3  bucket, using Whatsapp Token authentication, then converts the audio to text using the Amazon Transcribe  start_transcription_job  API, which leaves the transcript file in an Output Amazon S3 bucket.
Function that invokes audio_job_transcriptor looks like this:
Python
def start_job_transciptor (jobName,s3Path_in,OutputKey,codec):
    response = transcribe_client.start_transcription_job(
            TranscriptionJobName=jobName,
            IdentifyLanguage=True,
            MediaFormat=codec,
            Media={
            'MediaFileUri': s3Path_in
            },
            OutputBucketName = BucketName,
            OutputKey=OutputKey 
            )
💡 Notice that the IdentifyLanguage parameter is configured to True. Amazon Transcribe can determine the primary language in the audio.
  1. The  transcriber_done  Lambda Function is triggered with an  Amazon S3 Event Notification put item  once the Transcribe Job is complete. It extracts the transcript from the Output S3 bucket and sends it to  whatsapp_out  Lambda Function to respond to WhatsApp.
✅ You have the option to uncomment the code in the  transcriber_done  Lambda Function and send the voice note transcription to  langchain_agent_text  Lambda Function.
Python
try:       
    response_3 = lambda_client.invoke(
        FunctionName = LAMBDA_AGENT_TEXT,
        InvocationType = 'Event' ,#'RequestResponse', 
        Payload = json.dumps({
            'whats_message': text,
            'whats_token': whats_token,
            'phone': phone,
            'phone_id': phone_id,
            'messages_id': messages_id
        })
    )
    print(f'\nRespuesta:{response_3}')
    return response_3
    
except ClientError as e:
    err = e.response
    error = err
    print(err.get("Error", {}).get("Code"))
    return f"Un error invocando {LAMBDA_AGENT_TEXT}

3- LLM Processing:

LLM Processing

The agent receives the text and performs the following:
  1. Queries the Amazon DynamoDB table called  user_metadata  to see if the  session  has expired. If it is active, it recovers the  SessionID , necessary for the next step, if it expires it creates a new session timer.
  1. Queries the Amazon DynamoDB table called session Table to see if there is any previous conversation history.
  1. Consult the LLM through Amazon Bedrock using the following prompt:
Python
The following is a friendly conversation between a human and an AI. 
    The AI is talkative and provides lots of specific details from its context. 
    If the AI does not know the answer to a question, it truthfully says it does not know.
    Always reply in the original user language.
    Current conversation:
    {history}
    Human:{input}
    Assistant:
  1. Send the response to WhatsApp through  whatsapp_out  the Lambda Function.
💡 The phrase "Always reply in the original user language" ensures that it always responds in the original language and the multilingual capacity is provided by  Anthropic Claude , which is the model used in this application.

Let's build!

Step 0: Activate WhatsApp account Facebook Developers

1-  Get Started with the New WhatsApp Business Platform 
2-  How To Generate a Permanent Access Token — WhatsApp API 
3-  Get started with the Messenger API for Instagram 

Step 1: APP Set Up

✅ Clone the repo
git clone  https://github.com/build-on-aws/building-gen-ai-whatsapp-assistant-with-amazon-bedrock-and-python 
✅ Go to:
cd private-assistant

Step 2: Deploy architecture with CDK.

Architecture

In  private_assistant_stack.py  edit this line with the whatsapp Facebook Developer app number:
 DISPLAY_PHONE_NUMBER = 'YOU-NUMBER' 
This agent manages conversation memory, and you must set the session time  here  in this line:
 if diferencia > 240: #session time in seg 
Tip:  Kenton Blacutt , an AWS Associate Cloud App Developer, collaborated with Langchain, creating the  Amazon Dynamodb based memory class  that allows us to store the history of a langchain agent in an  Amazon DynamoDB .
✅ Create The Virtual Environment: by following the steps in the  README 
plaintext
python3 -m venv .venv
source .venv/bin/activate
for windows:
plaintext
.venv\Scripts\activate.bat
✅ Install The Requirements:
plaintext
pip install -r requirements.txt
✅ Synthesize The Cloudformation Template With The Following Command:
plaintext
cdk synth
✅🚀 The Deployment:
plaintext
cdk deploy
deployment_time


Step 3: WhatsApp Configuration

Edit WhatsApp configuration values in Facebook Developer in  AWS Secrets Manager   console .
Secret

✅ The verification token is any value, but it must be the same in step 3 and 4.

Step 4: Webhook Configuration

  1. Go to  Amazon API Gateway Console 
  1. Click on  myapi .
  1. Go to Stages -> prod -> /cloudapi -> GET, and copy Invoke URL.
  1. Configure Webhook in the Facebook developer application.
  • Set Invoke URL.
  • Set verification token.
Invoke Url



Enjoy the app!:

✅ Chat and ask follow-up questions. Test the app's abilities to handle multiple languages.

✅ Send and transcribe voice notes. Test the app's capabilities for transcribing multiple languages.


🚀 Keep testing the app, play with the prompt  langchain_agent_text  Amazon Lambda function and adjust it to your need.

Conclusion:

In this tutorial, you deployed a serverless WhatsApp application that allows users to interact with an LLM through Amazon Bedrock. This architecture uses API Gateway as a connection between WhatsApp and the application. Amazon Lambda functions process code to handle conversations. Amazon DynamoDB tables manage and store message information, session details, and conversation history.
You now have the essential code to improve the application. One option moving forward is to incorporate  Retrieval-Augmented Generation (RAG)  to generate more sophisticated responses depending on the context.
To handle customer service scenarios, the application could connect to  Amazon Connect  and transfer calls to an agent if the LLM cannot resolve an issue.
With further development, this serverless architecture demonstrates how conversational AI can power engaging and useful chat experiences on popular messaging platforms.

🚀 Some links for you to continue learning and building:

# generative-ai# amazon-bedrock# llm# lambda
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article