An Iron Man-inspired AI assistant with voice capabilities, built with Python and OpenAI.
- Voice Interaction: Talk to JARVIS and get voice responses
- Siri-like Voice Quality: Natural speech patterns and professional voice
- OpenAI Integration: Powered by GPT-3.5-turbo for intelligent responses
- Windows Native TTS: High-quality text-to-speech using Windows built-in voices
- Multiple Voice Options: Automatically selects the best available voice
- Natural Speech Processing: Pauses, emphasis, and natural flow like Siri
- Python 3.7+
- Windows 10/11 (for voice features)
- OpenAI API key
- WSL (Windows Subsystem for Linux) recommended
-
Clone the repository:
git clone https://github.com/yourusername/jarvis-ai.git cd jarvis-ai -
Install dependencies:
pip install -r requirements.txt
-
Configure your OpenAI API key: Edit
config.pyand add your OpenAI API key:OPENAI_API_KEY = "your-openai-api-key-here"
For a modern messenger-like interface with text and voice input:
python jarvis_gui.py
# or
python launch_jarvis_gui.pypython jarvis_windows.pypython jarvis.pypython jarvis_macos.py- Natural speech patterns with pauses and emphasis
- Professional male voice selection
- Multiple voice testing to find the best quality
- Emphasis on important words like "JARVIS", "Sir", "Tony"
- Rate: Slightly slower for clarity (-2)
- Volume: Full volume (100)
- Pitch: Natural pitch (0)
- Voice: Microsoft David Desktop (professional male)
- Adds natural pauses after greetings
- Emphasizes key words and names
- Includes dramatic pauses for questions
- Natural flow with comma pauses
jarvis-ai/
├── jarvis_windows.py # Main voice-enabled JARVIS (Siri-like)
├── jarvis.py # Basic voice-enabled JARVIS
├── config.py # Configuration settings
├── requirements.txt # Python dependencies
├── templates/ # Web interface templates
└── README.md # This file
Edit config.py to customize:
- OpenAI API key
- Voice settings
- User name
- Speech rate and volume
- Talk to JARVIS: Just speak or type your questions
- Exit: Say "goodbye", "quit", "exit", or "stop"
- Voice Input: Type your messages (voice input requires Windows Speech Recognition setup)
- Windows Speech Recognition (when enabled)
- Text input fallback for immediate use
- Natural language processing
- Windows native TTS with multiple voice options
- Siri-like speech patterns
- Professional voice quality
- Natural pauses and emphasis
- Powered by OpenAI GPT-3.5-turbo
- Siri-like personality
- Knowledgeable about any topic
- Concise and engaging responses
- Ensure Windows audio is working
- Check if Windows TTS is available
- Try running in text-only mode
- The system automatically tests multiple voices
- Falls back to default voice if needed
- Voice settings can be adjusted in code
- Check your API key in
config.py - Ensure you have OpenAI API credits
- Check internet connection
- Fork the repository
- Create a feature branch
- Make your changes
- Test thoroughly
- Submit a pull request
This project is open source and available under the MIT License.
- Inspired by Iron Man's JARVIS
- Voice design inspired by Siri
- Built with OpenAI GPT-3.5-turbo
- Uses Windows native text-to-speech
If you encounter any issues:
- Check the troubleshooting section
- Ensure all dependencies are installed
- Verify your OpenAI API key is correct
- Try running in text-only mode first
"Sometimes you gotta run before you can walk." - Tony Stark
JARVIS AI Assistant - Your personal AI companion