A multilingual, senior-friendly AI Voice Banking Assistant built using FastAPI, Streamlit, Speech-to-Text, Text-to-Speech, and Google Gemini. It includes reminders, SOS alerts, user profiles, and mock banking APIs.
- 🎤 Voice Assistant (STT → LLM → TTS)
- 🆘 SOS Emergency Alerts
- ⏰ Reminders Manager
- 🏦 Mock Banking Services
- 👤 User Profile
- ⚙️ Dark Mode & Accessibility Mode
- 🗂️ Dashboard Navigation with Streamlit
Frontend:
- Streamlit
- Python
Backend:
- FastAPI
- Uvicorn
- Google Gemini API
- SpeechRecognition
- gTTS / Pygame
git clone <your-repo-url>
cd GHCI-mainpython -m venv venv
venv\Scripts\activatepip install -r requirements.txtcd backend
uvicorn backend:app --reload --port 8000Backend runs at: 👉 http://127.0.0.1:8000
Open a new terminal (backend must stay running):
cd frontend
streamlit run app.pyFrontend opens at: 👉 http://localhost:8501
GHCI-main/
│── backend/
│ └── backend.py # FastAPI backend
│── frontend/
│ ├── app.py # Dashboard
│ ├── ui_theme.py # Theme manager
│ └── pages/
│ ├── Voice.py
│ ├── Reminders.py
│ ├── SOS.py
│ ├── Banking.py
│ ├── Profile.py
│ └── Settings.py
│── requirements.txt
│── .gitignore
│── README.md
Create a .env file inside backend/:
GEMINI_API_KEY=your_api_key_here