How I Built a WhatsApp Translation Bot Using Node.js

In a world where instant communication is key, language barriers can be a major challenge. So, I built a WhatsApp translation bot that automatically detects and translates messages between English and Malayalam. This bot uses Node.js, whatsapp-web.js, translate-google, and franc for language detection and translation.

In this article, I’ll walk you through how I built it — step by step — with the exact code I used. If you’re looking to create your own WhatsApp translator, this guide will help you get started.

🔹 Step 1: Installing the Required Dependencies

Before writing the code, we need to install a few essential Node.js libraries. Open your terminal and run:

npm install whatsapp-web.js qrcode-terminal translate-google franc

Here’s what each package does:

• whatsapp-web.js — Allows us to automate WhatsApp messages.
• qrcode-terminal — Displays a QR code for authentication.
• translate-google — Enables text translation using Google Translate.
• franc — Detects the language of a given text.

Once installed, we can move on to writing the bot’s logic.

🔹 Step 2: Writing the WhatsApp Bot Code

Now, let’s create a file called index.js and paste the following code:

const { Client, LocalAuth } = require('whatsapp-web.js');

const client = new Client({
    authStrategy: new LocalAuth({
        dataPath: '.wwebjs_auth'
    })
});

const qrcode = require('qrcode-terminal');
client.on('ready', () => {
    console.log('✅ Client is ready!');
});

client.on('qr', qr => {
    qrcode.generate(qr, { small: true });
});

client.initialize();

// Importing the translation and language detection modules
const translate = require('translate-google');

client.on('message_create', async (message) => {
    if (message.fromMe) return; // Ignore messages sent by the bot

    const franc = await import('franc').then(module => module.franc);
    let langCode = franc(message.body);

    // Assume English for short text if franc returns "und"
    if (langCode === 'und' && /^[A-Za-z0-9\s.,!?"'()]+$/.test(message.body)) {
        langCode = 'eng';
    }

    const malayalamCodes = ['ml', 'mal']; // Codes for Malayalam

    if (langCode === 'eng') {
        console.log("✅ Detected English. Translating to Malayalam...");
        try {
            const translatedText = await translate(message.body, { to: 'ml' });
            console.log("📌 Translation:", translatedText);
            await message.reply(translatedText);
        } catch (error) {
            console.error("❌ Translation error:", error);
            await message.reply("Sorry, I couldn't translate that.");
        }
    } else if (malayalamCodes.includes(langCode)) {
        console.log("✅ Detected Malayalam. Translating to English...");
        try {
            const translatedText = await translate(message.body, { to: 'en' });
            console.log("📌 Translation:", translatedText);
            await message.reply(translatedText);
        } catch (error) {
            console.error("❌ Translation error:", error);
            await message.reply("Sorry, I couldn't translate that.");
        }
    } else {
        console.log("🚫 Unsupported language detected.");
    }

    console.log("🔍 Detected language code:", langCode);
});

💡 Breaking Down the Code:

1. Authentication: Uses whatsapp-web.js with local storage (LocalAuth) to keep the session active.
2. QR Code Login: On first-time setup, a QR code appears in the terminal. Scan it with WhatsApp to link the bot.
3. Message Listener: The bot listens for new messages and ignores its own messages to avoid infinite loops.
4. Language Detection: The franc library determines whether the message is in English or Malayalam.
5. Translation Logic:

• English → Malayalam
• Malayalam → English
• Other languages are ignored (but can be expanded).

6. Error Handling: If translation fails, the bot informs the user.

🔹 Step 3: Running the Bot for the First Time

Now, it’s time to test our bot! Simply run:

node index.js

If it’s your first time, a QR code will appear in the terminal. Open WhatsApp on your phone, go to Linked Devices, and scan the code.

Once scanned, the bot will start running and responding to messages.

🔹 Step 4: Improving the Bot (Optional Enhancements)

Now that the bot works, here are a few ways to improve it:

✅ Support More Languages: Instead of only English and Malayalam, add more languages dynamically.
✅ Deploy on a VPS: Host it on a server so it runs 24/7.
✅ Use a More Advanced Language Detector: Replace franc with fasttext for better accuracy.

🔹 Final Thoughts

Building this WhatsApp translation bot was a great experience, and I’ve made the complete code available on my GitHub. You can check it out here: GitHub — FAWAZMALIK/whatsapp-translation-bot.

Feel free to explore, modify, and improve it! 🚀 If you have any suggestions or want to contribute, drop by the repository and share your thoughts.