🌐 A smart, AI-powered assistant that understands text, voice, images, and PDFs.
🤖 Built using Mistral LLM + Streamlit.
🎬 Check out a quick walkthrough of the app below:
- ✅ Multimodal Input — Accepts text, speech, images, and PDFs
- 🧠 Domain-Specific AI — Tailored responses for Education, Healthcare, etc.
- 🖥️ Responsive UI — Sidebar navigation, loading animations, dark mode–ready
- ⚡ Real-time Response — Powered by Mistral LLM API
| Tool | Role |
|---|---|
| Python | Backend logic |
| Streamlit | UI & Web App Framework |
| Mistral API | LLM for response generation |
| SpeechRecognition | Voice input support |
| PyTesseract | OCR from images |
| PyPDF2 | PDF text extraction |
# 1. Clone the repository
git clone https://github.com/yourusername/multimodal-chatbot.git
cd multimodal-chatbot
# 2. Create virtual environment
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Add MISTRAL API key to `.streamlit/secrets.toml`
[mistral]
api_key = "YOUR_MISTRAL_API_KEY"
# 5. Run the application
streamlit run app.pyYou can easily deploy this app to any modern cloud platform:
- 🌐 Render
- 🚀 Streamlit Cloud
- 📦 Heroku
🔑 Note: Always store sensitive credentials securely using environment variables or
.streamlit/secrets.toml.
Here's what's planned:
- 🧩 Plugin System — Modular domain expansion
- 💬 Conversational Memory — Context retention
- 🌍 Multilingual Support
- 🔐 Robust Error Handling
Licensed under the MIT License — see the LICENSE file.
Adyaprana Pradhan
- GitHub: https://github.com/Adyaprana
- LinkedIn: https://www.linkedin.com/in/adyaprana21
- 🌐 Website: adyaprana.com
- 🐦 Twitter: @Adyaprana
- 📸 Instagram: @adyaprana