WhatsAppBot

This is the bot that can. Powered by FastAPI and WhatsApp Business Cloud API.

Features

• Intelligent Text and Image Input: Ask complex questions, get creative prompts answered, and explore different styles with Google's cutting-edge language models.
• Seamless WhatsApp Integration: Send and receive messages through WhatsApp, making AI accessible and convenient right at your fingertips.
• Huggingface Inference Support (Beta): Access a broad range of pre-trained models for diverse text tasks while exploring the potential of Huggingface's powerful ecosystem.

Getting Started

Prerequisites

• Python 3.10.13
• FastAPI
• WhatsApp Business Cloud API credentials
• Google Cloud Platform (GCP) credentials/service account with access to Vertex AI (for Google LLMs)
• (Optional) Additional APIs specific to your desired functionality

Installation

Option 1: Easy Setup (Recommended)

1. Download the repository

   git clone https://github.com/so-dipe/WhatsAppBot.git
   cd WhatsAppBot

2. Run the installation script:

   ./install.sh

Option 2: Manual Setup:,

1. Create a Python Virtual Environment in the repository:

   python -m venv venv

2. Install dependencies in the virtual environment:

   source venv/bin/activate
   pip install -r requirements.txt

3. Configure settings and Environment Variables

   1. Copy the sample_config.env file

      cp sample_config.env .env

   2. Open .env and replace the placeholder values with your credentials

   3. If you plan on using google's models, paste a copy of your service account at service_account_key.json

   alternatively, run the configure.sh script

   ./configure.sh

Usage

1. Run run_app.sh bash script

   ./run_app.sh

OR Manually start FastAPI and Redis servers.

1. Run FastAPI server

   uvicorn app.main:app 

   in developement mode, use

   uvicorn app.main:app --reload

   by default, a port is opened on 8000.

2. Run Redis Server

   redis-server config/redis.conf

   config/redis.conf contains the default config file used.

3. Set up ngrok (if needed) to expose your local server to the internet:

   ngrok http 8000

   Note: WhatsApp Business API requires a webhook callback URL that must be accessible from the internet. Explore ngrok for a permanent domain if you prefer.

Contributing

Roadmap

• Seamless Huggingface Inference API Integration: Empower users with a vast library of pre-trained models and a user-friendly model selection interface.
• Expand LLM Hosting Platform Support: Welcome DeepInfra, AnyScale, and potentially more to offer diverse LLM hosting options.
• Unleash Speech-to-Text Capabilities: Enable voice-based interactions for a more natural and accessible experience.
• Unlock the Power of Documents: Facilitate conversations around PDFs and document files for enriched knowledge sharing.

License

This project is licensed under the GNU General Public License v3.0. See the LICENSE file for details.