🩺 VaidyaVerse - AI Powered Bilingual Doctor Assistant

VaidyaVerse is an AI-powered medical assistant that analyzes voice input and medical images to generate doctor-like responses in Hindi and English.

The system integrates Speech-to-Text, Large Language Models, and Text-to-Speech to simulate a conversational AI doctor experience.

---

🚀 Features

• 🎤 Speech-to-Text using Groq Whisper
• 🧠 Medical reasoning using Groq LLM
• 🖼 Image-based medical query support
• 🌍 Supports Hindi (Devanagari) and English
• 🔁 Automatic Urdu-to-Hindi script normalization
• 🔊 Text-to-Speech using:
  • gTTS (Google Text-to-Speech)
  • ElevenLabs (Optional High-Quality Voice)
• ⚡ Interactive UI built with Gradio

---

🛠 Tech Stack

• Python
• Groq API (Whisper + LLM)
• ElevenLabs API (TTS)
• gTTS
• Gradio
• Indic Transliteration
• dotenv

---

📦 Installation

Clone the repository:

git clone https://github.com/YOUR_USERNAME/VaidyaVerse.git
cd VaidyaVerse

Install dependencies:

pip install -r requirements.txt

Create a .env file in the root directory:

GROQ_API_KEY=your_groq_api_key
ELEVENLABS_API_KEY=your_elevenlabs_api_key

Run the application:

python gradio_app.py

---

🌍 Language Support

Supported Languages:

• Hindi (Devanagari script)
• English

Behavior:

• If Hindi speech is detected → Hindi response + Hindi voice
• If English speech is detected → English response + English voice
• Urdu script is automatically normalized to Hindi

No other languages are supported.

---

🔊 Text-to-Speech Options

1️⃣ gTTS

• Lightweight
• Free
• Suitable for basic Hindi and English voice output

2️⃣ ElevenLabs (Optional)

• High-quality natural voice
• Better pronunciation
• Requires API key
• Internet connection required

---

⚠ Limitations

• This project is for educational and demonstration purposes only.
• It is NOT a substitute for professional medical advice.
• AI-generated responses may be incorrect or incomplete.
• No real medical validation layer is implemented.
• Image analysis is prompt-based and not a true computer vision diagnosis system.
• Requires active internet connection.
• Depends on Groq API uptime and rate limits.
• No patient data storage or authentication system.
• Not suitable for emergency or critical medical decisions.

---

🔒 Security Notice

• Never upload your .env file.
• Keep API keys private.
• Do not expose keys in commits.
• Use environment variables in production deployment.

---

📌 Future Improvements

• Add language selection toggle in UI
• Add medical safety guardrails
• Add disclaimer popup in UI
• Add conversation memory
• Deploy to HuggingFace Spaces
• Add authentication system
• Integrate real computer vision model
• Add structured medical output format

---

📄 License

This project is open-source and intended for educational use only.

---

👨‍💻 Author

Developed as an AI bilingual medical assistant project using Groq and ElevenLabs integration.