🎙️ Voice2Query AI

AI-Powered Speech-to-SQL and Text-to-SQL Database Assistant for University Database Exploration.

Voice2Query AI enables users to interact with relational databases using natural language voice commands or text queries. The system automatically converts user input into executable SQL queries, retrieves data from a PostgreSQL database, and visualizes results through an interactive Streamlit dashboard.

---

🚀 Features

• 🎤 Voice-to-Text Query Processing
• ⌨️ Natural Language Text Queries
• 🤖 AI-Powered SQL Generation using Ollama (Llama 3)
• 🗄️ PostgreSQL Database Integration
• 📊 Interactive Dashboard and Visual Analytics
• 🔍 Automatic Query Execution
• 📈 Charts and Data Insights
• 🔐 Environment-Based Configuration

---

🏗️ System Architecture

Voice Input / Text Input
            │
            ▼
 Speech-to-Text Processing
            │
            ▼
 Natural Language Query
            │
            ▼
 Ollama (Llama 3)
 Text-to-SQL Generation
            │
            ▼
 SQL Validation
            │
            ▼
 PostgreSQL Database
            │
            ▼
 Streamlit Dashboard
            │
            ▼
 Results & Visualizations

---

🛠️ Technology Stack

Component	Technology
Programming Language	Python
Database	PostgreSQL
Frontend Dashboard	Streamlit
LLM Engine	Ollama (Llama 3)
Data Processing	Pandas
Visualizations	Plotly
ORM / Database Layer	SQLAlchemy
Speech Processing	Speechmatics (Planned)

---

📂 Project Structure

Voice2Query-AI/
│
├── app/
│   ├── dashboard/
│   ├── database/
│   ├── nlp/
│   ├── speech/
│   ├── utils/
│   └── config.py
│
├── requirements.txt
├── dashboard.py
├── .env.example
├── README.md
└── .gitignore

---

⚙️ Installation

1. Clone Repository

git clone https://github.com/Madhumithra1321/Voice2Query-AI.git
cd Voice2Query-AI

2. Create Virtual Environment

python -m venv venv

Activate:

Windows:

venv\Scripts\activate

Linux/Mac:

source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

---

🗄️ Database Setup

Create a PostgreSQL database named:

voice2query

Initialize sample tables:

python -m app.database.database_setup

---

🔧 Configuration

Create a local environment file:

Copy-Item .env.example .env

Update the values inside .env.

PostgreSQL

DATABASE_URL=postgresql+psycopg2://username:password@localhost:5432/voice2query

Ollama (Default)

TEXT_TO_SQL_PROVIDER=ollama
OLLAMA_MODEL=llama3:latest
OLLAMA_URL=http://localhost:11434/api/generate
---

## ▶️ Running the Application

Start the dashboard:

```bash
streamlit run dashboard.py

Open:

http://localhost:8501

---

📊 Example Queries

• Show top 10 students by marks
• List all professors
• Display students from Naples
• Find highest scoring students
• Show courses with more than 5 credits

👥 Done by

• Madhumithra Balasubramanian
• Aya Hechaichi
• Alina Siddiqui

University of Naples Federico II