How to Integrate Microsoft Azure Cognitive Services with WhatsApp

Set Up Microsoft Azure Cognitive Services

• Go to the Microsoft Azure portal and sign in with your Azure credentials.

 

• Search for "Cognitive Services" in the search bar and select it from the list.

 

• Click on "Create" to start a new instance. Fill in the required details like Subscription, Resource group, and the name of your service.

 

• Choose the type of Cognitive Service you want to use (e.g., Text Analytics, Language Understanding). Note the API key and endpoint URL as these will be used later.

Set Up WhatsApp API

• Sign up for a service that provides a WhatsApp Business API, such as Twilio or MessageBird, as you can’t directly interact with WhatsApp from Azure.

 

• Configure your WhatsApp Business account through the service provider’s console, ensuring that you have the API key and URL for sending messages.

Create a Bridge Service

• Establish a local or cloud-based application that acts as a bridge between Azure Cognitive Services and the WhatsApp API. You can use Node.js or Python as the programming language for handling HTTP requests and responses.

 

• Use the following Node.js code as an example to set up your project:

const express = require('express');
const axios = require('axios');
const app = express();

app.use(express.json());

const azureCognitiveServiceEndpoint = 'YOUR_AZURE_SERVICE_ENDPOINT';
const azureApiKey = 'YOUR_AZURE_API_KEY';
const whatsappEndpoint = 'YOUR_WHATSAPP_ENDPOINT';
const whatsappApiKey = 'YOUR_WHATSAPP_API_KEY';

app.post('/webhook', async (req, res) => {
    try {
        const message = req.body.message; // WhatsApp message
        const cognitiveResponse = await axios.post(azureCognitiveServiceEndpoint, {body: message}, {
            headers: {'Ocp-Apim-Subscription-Key': azureApiKey}
        });

        const analyzedMessage = cognitiveResponse.data;
        await axios.post(whatsappEndpoint, {
            headers: {'Authorization': `Bearer ${whatsappApiKey}`},
            message: analyzedMessage
        });

        res.status(200).send('Message sent to WhatsApp');
    } catch (error) {
        console.error('Error processing request:', error);
        res.status(500).send('Error');
    }
});

const PORT = process.env.PORT || 3000;
app.listen(PORT, () => {
    console.log(`Server running on port ${PORT}`);
});

Deploy and Test the Service

• Deploy your application to a cloud service like Heroku, AWS, or Azure App Services to make it publicly accessible.

 

• Set up a webhook through your WhatsApp API provider’s dashboard. Ensure this webhook URL points to your deployed application.

 

• Send a test message from your WhatsApp to verify that the Cognitive Service processes it correctly and sends a response back via WhatsApp.