An intelligent conversational AI chatbot for the Urdu language, enabling natural, human-like voice interactions using Gemini LLM and Streamlit.
🔗 Live App: urduvoicebot.streamlit.app
🎥 Demo Video: Watch on YouTube
- 🎙️ Voice-to-Text in Urdu using Google Speech Recognition API
- 💬 Natural Language Understanding via Gemini LLM
- 🔊 Text-to-Speech (TTS) in Urdu using Google Text-to-Speech API
- 🌐 Streamlit Interface for interactive use
- 🧠 Optional fine-tuning of LLM for better Urdu understanding
-
Speech Recognition → Convert user's Urdu voice input into text using Google Speech API.
-
Language Processing with Gemini → Feed the text to the Gemini LLM for generating a relevant response.
-
Text-to-Speech Conversion → Convert Gemini's Urdu text response into audio using Google TTS API.
-
Streamlit App UI → Everything runs in an interactive Streamlit web interface.
Ensure you have the following installed:
- Python 3.8 or above
pip(Python package manager)- API key for Gemini (from Google AI Studio)
-
Clone the Repository
git clone https://github.com/Mubeen-Fatima/urdu-voice-chatbot.git cd urdu-voice-chatbot -
Create a Virtual Environment
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate
-
Install Dependencies
pip install -r requirements.txt
-
Configure Secrets
This project uses Streamlit secrets management. Create a
.streamlit/secrets.tomlfile:[gemini] GEMINI_API_KEY = "your_gemini_api_key"
-
Run the App
streamlit run app.py
urdu-voice-chatbot/
├── voicechatbot.py # Main Streamlit app
├── requirements.txt # Python dependencies
└── .streamlit/
└── secrets.toml # API key configurationWebsite: mubeenf.com
LinkedIn: linkedin.com/in/mubeenf
Upwork: upwork.com/freelancers/mubee
Feel free to fork, contribute, or reach out for collaboration!