🔊 JARVIS: A Voice-Controlled Virtual Assistant in Python

📌 Project Overview

JARVIS is a voice-controlled AI-based desktop assistant built using Python. Inspired by Iron Man’s J.A.R.V.I.S, this assistant performs real-time voice recognition, executes web-based commands, speaks results aloud using TTS, fetches live news, plays songs via voice commands, and integrates with OpenAI's GPT model to provide intelligent conversational capabilities.

---

🧠 Features

• 🎤 Voice Wake Word Detection – Listens for “Jarvis” to activate.
• 🌐 Web Automation – Opens Google, YouTube, Facebook, LinkedIn via command.
• 📰 Live News Updates – Fetches and speaks top headlines from NewsAPI.
• 🎵 Play Music – Plays songs using pre-defined music dictionary.
• 💬 Chat with GPT-3.5 Turbo – Handles intelligent queries via OpenAI API.
• 🔊 Text-to-Speech – Responds using Google TTS and Pyttsx3.
• 🎮 Multimedia Playback – Uses Pygame mixer for MP3 audio handling.

---

🛠️ Tech Stack

Module	Purpose
speech_recognition	Voice input and recognition
pyttsx3	Text-to-speech (offline)
gTTS	Google Text-to-Speech (online)
pygame	Audio playback of MP3
webbrowser	Browser automation
requests	API requests (News API, etc.)
openai	GPT-3.5 chat interface
pyaudio	Microphone access (via PortAudio)

---

📂 Project Structure

jarvis/
├── main.py               # Main program file
├── musicLibrary.py       # Contains music dictionary
├── client.py             # (Optional) OpenAI helper script
├── requirements.txt      # Dependencies
└── README.md             # Project overview (this file)

---

🚀 Setup Instructions

1. 🖥️ System Requirements

• OS: macOS / Windows / Linux
• Python ≥ 3.8
• Working microphone
• Internet connection (for TTS and API calls)

2. 📦 Install Dependencies

Install using pip:

pip install -r requirements.txt

If pyaudio fails on macOS, run:

brew install portaudio
pip install pyaudio

3. 🔑 Configure API Keys

Replace placeholder values with your actual keys in main.py:

newsapi = "<Your News API Key>"
client = OpenAI(api_key="<Your OpenAI API Key>")

Sign up for keys from:

• NewsAPI: https://newsapi.org/
• OpenAI: https://platform.openai.com/

---

🔊 How to Run

python main.py

Then say "Jarvis" to activate, and speak your command after the response.

---

🧪 Example Commands

• “Jarvis → open google”
• “Jarvis → play despacito”
• “Jarvis → tell me the latest news”
• “Jarvis → what is the capital of Japan?”
• “Jarvis → open YouTube”

---

🧠 Limitations

• Works best in a quiet environment.
• Internet required for APIs (News, OpenAI, gTTS).
• Wake-word detection is primitive (basic string match).
• No deep NLP understanding without GPT API.

---

📚 Academic Purpose

This project was built as an academic showcase of practical voice-based AI integration using Python libraries and APIs. It demonstrates:

• Use of AI/ML APIs (OpenAI GPT)
• Real-time audio processing
• API integration and error handling
• Voice user interface (VUI) design

---

🙏 Acknowledgements

• OpenAI for GPT-3.5 Turbo
• Google for gTTS
• NewsAPI.org for real-time news access
• SpeechRecognition contributors
• Python Community for rich open-source support

---

🧑‍💻 Developed By

Ekaagra gupta B.Tech CSE (AI & ML) Manipal University Jaipur Email work: \ekaagra.2427010368@muj.manipal.edu Email personal: \ekaagrag2006@gmail.com GitHub:\https://github.com/ekaagragupta

---

📄 License

This project is for educational and academic use only. Not for commercial deployment.