🦷 WhatsApp Dental Appointment Bot

This project implements a WhatsApp-based AI assistant for dental clinics, enabling patients to book appointments via text or voice messages. It leverages LangGraph for conversational flows, FastAPI for the backend, and integrates with Google Calendar for scheduling.

🧠 Features

• Multi-Modal Communication: Handles both text and voice messages.
• Voice Transcription: Transcribes audio messages using the Whisper model via Groq.
• Conversational AI: Processes user inputs through LangGraph-powered chatbot.
• Google Calendar Integration: Checks for available slots and creates appointments.
• FastAPI Backend: Provides API endpoints for WhatsApp webhook and other functionalities.
• Project Management with uv: Utilizes uv for dependency management and project setup.

🚀 Getting Started

Prerequisites

• Python 3.12 or higher
• uv installed globally

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/whatsapp-dental-bot.git
   cd whatsapp-dental-bot

2. Initialize the project with uv:

   uv init --app

3. Add dependencies:

   uv add fastapi langgraph langchain langchain_groq google-api-python-client python-dotenv

4. Set up environment variables:

   Create a .env file in the root directory and add the following:

   GROQ_API_KEY=your_groq_api_key
   OPENAI_API_KEY=oepnai_api_key
   FACEBOOK_AUTH_TOKEN=facebook_auth_token
   SENDER_PHONE=sender_phone

Ensure you have the necessary credentials for Google Calendar API access.

5. Run the application:

   uv run main.py

The application will be available at http://127.0.0.1:5001.

🛠️ LangGraph Flow

The conversational flow is managed by LangGraph with the following logic:

• Text Message: Directly processed by the chatbot node.
• Audio Message: Routed to the audio node for transcription, then passed to the chatbot node.
• Chatbot Node: Interacts with the tools node to perform actions like checking and creating appointments in Google Calendar.

This modular design allows for easy scalability and maintenance.

📅 Google Calendar Integration

The tools node utilizes the Google Calendar API to:

• Check Availability: Search for free time slots.
• Create Appointments: Schedule new appointments based on user input.

Setting Up Google Calendar API Credentials

1. Create a Google Cloud Project:

   • Go to the Google Cloud Console.
   • Click on Select a project and then New Project.
   • Enter a project name and click Create.(support.google.com)

2. Enable Google Calendar API:

   • In the APIs & Services Dashboard, click + ENABLE APIS AND SERVICES.
   • Search for "Google Calendar API" and click on it.
   • Click Enable to activate the API for your project.(endgrate.com)

3. Configure OAuth Consent Screen:

   • Navigate to APIs & Services > OAuth consent screen.
   • Choose External or Internal based on your needs and click Create.
   • Fill out the required fields such as App name, User support email, and Developer contact information.
   • Click Save and Continue through the scopes and test users sections.(endgrate.com, developers.google.com)

4. Create OAuth 2.0 Credentials:

   • Go to APIs & Services > Credentials.
   • Click + CREATE CREDENTIALS and select OAuth client ID.
   • Choose Desktop app as the application type.
   • Enter a name and click Create.
   • Click Download JSON to obtain your credentials.json file.(developers.google.com, support.google.com, developers.google.com)

5. Place credentials.json in Your Project:

   • Move the downloaded credentials.json file to the root directory of your project.
   • Ensure this file is included in your .gitignore to prevent it from being committed to version control.

6. Authenticate and Generate token.json:

   The first time you run your application, it will prompt you to authenticate with your Google account. Upon successful authentication, a token.json file will be created, storing your access and refresh tokens.

   Note: The token.json file is user-specific and should also be excluded from version control.

For more detailed instructions, refer to the Python Quickstart Guide for Google Calendar API.

🎤 Audio Transcription with Whisper

Audio messages received from WhatsApp are processed as follows:

1. Reception: Audio message is received via the FastAPI endpoint.
2. Transcription: The audio node uses the Whisper model via Groq to transcribe the audio into text.
3. Processing: Transcribed text is then passed to the chatbot node for further handling.

This ensures seamless handling of both text and voice inputs.

📞 WhatsApp Integration via Meta

To integrate WhatsApp with your application, follow these steps:

1. Create a Meta Developer Account

• Visit Meta for Developers and log in with your Facebook account.
• Accept the terms and set up your developer profile.

2. Create a Meta App

• Navigate to My Apps and click on Create App.
• Select Other as the app type and click Next.
• Choose Business as the app type and click Next.
• Enter an App Name, provide a Contact Email, and select your Business Account (if available). Click Create App.

3. Add WhatsApp Product to Your App

• In your app's dashboard, scroll down to the Add Products to Your App section.
• Find WhatsApp and click Set Up.
• Follow the on-screen instructions to set up the WhatsApp Business API.

4. Configure WhatsApp Business Account

• In the WhatsApp section of your app dashboard, navigate to Getting Started.

• Select or create a WhatsApp Business Account (WABA).

• Add and verify a phone number to be used for sending messages.

• Generate a Permanent Access Token with the necessary permissions:

  • whatsapp_business_messaging
  • whatsapp_business_management

5. Set Up Webhooks

• In your app dashboard, go to Webhooks and click Set Up Webhooks.

• Select the WhatsApp Business Account.

• Provide a Callback URL (your FastAPI endpoint) and a Verify Token.

• Subscribe to the following fields:

  • messages
  • message_deliveries
  • message_reads

6. Test Your Setup

• Use the Test Number provided in the Getting Started section to send