Aura: AI-Powered Voice Assistant using Python

Aura is a smart, voice-activated virtual assistant built using Python. Think of it as your personal desktop AI — capable of responding to voice commands, answering general questions via GPT, playing music, fetching weather updates, reading the latest news, and opening popular websites. Inspired by virtual assistants like Alexa, Siri, and Google Assistant, Nexa brings AI-powered interaction to your terminal.

---

What Can Aura Do?

• Voice-activated hotword detection ("Aura")
• Answer your questions using ChatGPT (GPT-3.5 Turbo)
• Search summaries from Wikipedia
• Play music using a YouTube-based music library
• Get current weather updates from any city
• Fetch the latest news headlines (from India)
• Open common websites like Google, YouTube, LinkedIn, etc.
• Tell you the current time and date

---

Tech Stack

Category	Tool/Library Used
Voice Input	speech_recognition
Text-to-Speech	gTTS, pyttsx3, pygame
AI Responses	openai (ChatGPT)
Music Search	fuzzywuzzy (string match)
Web Data	wikipedia, requests
News API	NewsAPI.org
Weather API	OpenWeatherMap
Browser Automation	webbrowser

---

Project Structure

nexa-voice-assistant/
├── main.py             # Core logic of Nexa assistant
├── client.py           # Standalone ChatGPT test client
├── musicLibrary.py     # Custom music-to-YouTube mapping
├── requirements.txt    # Python dependencies
└── README.md           # You're here!

---

Setup Instructions

1. Clone the Repository

git clone https://github.com/your-username/aura-voice-assistant.git
cd aura-voice-assistant

2. Install Required Dependencies

Install all dependencies using pip:

pip install -r requirements.txt

3. Add Your API Keys

Open main.py and client.py and update the placeholders:

# main.py
newsapi = "your_newsapi_key_here"
OpenAI(api_key="your_openai_api_key_here")
api_key = "your_openweathermap_key_here"

# client.py
client = OpenAI(api_key="your_openai_api_key_here")

How to Run

1. Run the Main Script

python main.py

2. Say the Hotword

Wait until you see:

Listening...

Say:

Aura

The assistant will respond with:

Ya

Now you're in command mode. Speak your command. Examples:

• "What is Machine Learning?"

• "Play Skyfall"

• "Weather in Bangalore"

• "News"

• "Open YouTube"

• "What is Python?"

Example Voice Commands

Command	What Happens
"Open Google"	Launches Google in your browser
"Play Skyfall"	Opens corresponding YouTube song
"Weather in Mumbai"	Fetches weather from OpenWeatherMap
"News"	Reads latest 5 headlines from NewsAPI
"What is AI?"	Answers using Wikipedia or GPT fallback
"Time" / "Date"	Announces current time or date

Custom Music Library

The music library is in musicLibrary.py. Add your favorite tracks like this:

music = {
    "skyfall": "https://youtube.com/your-skyfall-link",
    "stealth": "https://youtube.com/your-stealth-link"
}

API Key Security(Optional but highly recommended)

To keep your keys safe, do this:

1. Install python-dotenv:

pip install python-dotenv

2. Create a .env file in your root folder:

OPENAI_API_KEY=your_openai_key
NEWS_API_KEY=your_newsapi_key
WEATHER_API_KEY=your_openweathermap_key

3. In your code, update API key loading:

from dotenv import load_dotenv
load_dotenv()

import os
openai_key = os.getenv("OPENAI_API_KEY")

How It Works (Behind the Scenes)

1. Listening Loop: Constantly waits for the word "Aura".
2. Voice Activation: Once heard, it activates and listens again for your actual command.
3. Command Parsing: Determines intent (open site, play music, ask GPT, etc.).
4. Task Execution: Launches the relevant function:

• webbrowser.open() for URLs
• wikipedia.summary() for info
• openai.ChatCompletion for AI replies
• requests.get() for weather/news APIs

Testing GPT Separately

You can run client.py to test if your OpenAI API key works:

python client.py

Requirements

speechrecognition
pyttsx3
gtts
pygame
requests
wikipedia
fuzzywuzzy
openai
python-Levenshtein

Install them via:

pip install -r requirements.txt

Future Improvements

• Voice wake word detection using Vosk or Snowboy

• Whisper API for improved speech-to-text accuracy

• GUI interface using Tkinter or Streamlit

• Spotify/Youtube API for dynamic music search

• Context-based conversations via AI memory

• Support for multiple languages

Author

Developed by Aditya Dasappanavar
GitHub: AdityaD28

License

This project is licensed under the MIT License.

You’re free to modify, distribute, and use this software with attribution.