NIELS - Virtual Assistant Prototype

NIELS is an AI-powered virtual assistant named after Niels Bohr. This prototype version supports both voice and text-based interactions, assisting users with various tasks like opening applications, making calls, generating AI responses, and more. This is my first step to learn about Large Language Modela(LLM's) and their working.

---

🚀 Features

✔️ Voice & Text Chat Support
✔️ Application Control (Open apps with commands)
✔️ Call Making
✔️ AI Responses
✔️ Custom Hotword Activation (.ppn file for wake-word detection)

---

📂 Folder Structure

📆 Project Folder  
 ┓ 📂 frontend/      # UI and client-side code  
 ┓ 📂 backend/       # Server-side logic  
 ┓ 📂 notifs/        # Notifications-related functions  
 ┓ 📝 chat.html      # Chat interface  
 ┓ 📝 main.py        # Core functionality  
 ┓ 📝 run.py         # Execution script

---

🛠️ Installation

Prerequisites

• Python 3.10.10
• Required dependencies (listed in requirements.txt)

Steps

1️⃣ Clone the repository:

git clone https://github.com/JayanthSrinivas06/Virtual_AI_assistant.git
cd Virtual_AI_assistant

2️⃣ Create a virtual environment:

python -m venv env
source env/bin/activate  # (Windows: env\Scripts\activate)

3️⃣ Install dependencies:

pip install -r requirements.txt

4️⃣ Run the assistant:

python run.py

---

📸 Screenshots

Here’s a preview of the assistant in action:

Home Page: Voice command Interface:

🖥️ Tech Stack

• Frontend: HTML, CSS, JavaScript
• Backend: Python, eel
• Database: SQLite
• Hotword Detection: Porcupine .ppn file

---

🤖 LLM Used

NIELS leverages Cohere’s Command R LLM and Groq’s ultra-fast inference API to generate intelligent and context-aware AI responses.

Cohere’s Command R delivers robust multilingual natural language understanding and generation tailored for assistant tasks and reasoning.

Groq’s LPU-backed infrastructure enables ultra-low-latency and high-throughput LLM inference, ensuring real-time response quality even during complex conversations.
This dual-integration enhances both the cognitive capability and execution efficiency of NIELS in voice and text-based use cases.

---

👥 Contributors

Name	GitHub Profile
Jayanth Srinivas Bommisetty	JayanthSrinivas06
Sarvan Dattu Perumalla	Sarvan05
Kesava Parepalli	KesavaParepalli
Aryan Puri	Aryan-Puri-23

---

🌐 Links

• Project Repository: GitHub Link

---

⚠️ Note

This is a limited task prototype of NIELS. Improved versions with new features and advanced NLP capabilities are under development! 🚀