AI Voice Assistant

This project is an advanced AI voice assistant powered by Google's Gemini language model. It uses voice recognition to detect a wake word, processes user queries, and responds using text-to-speech technology.

Features

• Wake word detection ("Jarvis")
• Voice input processing
• Natural language understanding and generation using Gemini AI
• Text-to-speech output
• Conversation history management
• Integration with various tools:
  • Setting timers and alarms
  • Calendar management (Google Calendar API)
  • Weather information retrieval (OpenWeatherMap API)
  • News fetching (Google News API)
  • Smart home device control (Home Assistant API)
  • To-do list and reminder management
  • Web search functionality (Google Search API)
  • Message and email sending
  • Music and podcast playback
  • Directions and traffic information (Google Maps API)
  • Flight status checking
  • Phone messaging via Discord API

Requirements

• Python 3.8+
• pip
• virtualenv

Setup

1. Clone the repository:

   git clone https://github.com/xenioxyt/jarvis-gemini.git
   cd jarvis-gemini
   

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`
   

3. Install the required packages:

   pip install -r requirements.txt
   

4. Set up environment variables: Create a .env file in the project root and add the following:

   PICOVOICE_ACCESS_KEY=your_picovoice_access_key
   GEMINI_API_KEY=your_gemini_api_key
   GOOGLE_CREDENTIALS=path/to/your/google-credentials.json
   GOOGLE_API_KEY=your_google_api_key
   GOOGLE_CSE_ID=your_google_cse_id
   OPENWEATHER_API_KEY=your_openweather_api_key
   GOOGLE_CLOUD_API_KEY=your_google_cloud_api_key
   NEWS_API_KEY=your_news_api_key
   DISCORD_BOT_TOKEN=your_discord_bot_token
   

5. Ensure you have the necessary credentials file for Google Cloud services.

Usage

To start the voice assistant, run:

python main.py

The assistant will listen for the wake word "Jarvis". Once detected, it will record your query, process it using the Gemini AI, and respond using text-to-speech.

Project Structure

• main.py: The entry point of the application.
• wake_word_detector.py: Handles wake word detection.
• audio_recorder.py: Manages audio recording after wake word detection.
• gemini_api.py: Interfaces with the Gemini AI for natural language processing.
• tts_engine.py: Handles text-to-speech conversion and audio playback.
• tools.py: Contains various tool functions for extended functionality.
• prompt.py: Defines the system prompt for the AI assistant.
• function_tools/: Directory containing individual tool implementations:
  • weather.py: Retrieves weather information.
  • google_search.py: Performs web searches.
  • news.py: Fetches news articles.
  • directions.py: Provides directions and traffic information.
  • discord_message.py: Sends messages via Discord.
  • phone_message.py: Sends messages to phones.
  • place_info.py: Retrieves information about places.
  • take_notes.py: Manages note-taking functionality.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License.