SignLangAI is a Django-based web application that translates Russian text into videos with gestures of the Russian Sign Language (RSL), using pre-recorded gesture clips.
The project is designed for learning and communication with a limited set of gestures.
- ✅ Translate phrases into a sequence of available gestures
- 🎥 Automatically merge gesture clips into a single video
- 💬 Chat history for authenticated users
- 🔐 User registration and authentication
- 🧠 GPT-4 for semantic phrase analysis and simplification to available gestures
- Backend: Django, MoviePy, GPT-4 (via gpt4free)
- Frontend: HTML + JavaScript + jQuery (via Django templates)
- Database: SQLite (default)
- Authentication: Django auth
- Video Storage:
media/gestures/— individual gesture clipsmedia/outputs/— generated videos
- User enters a phrase.
- GPT-4 analyzes the phrase and generates a list of
.mp4gesture files. - The program merges these clips into a single video.
- The final video is displayed in the chat window.
Input:
Привет, как ты?
GPT-4 Output:
['привет.mp4', 'как.mp4', 'ты.mp4']
Result:
A stitched video with consecutive gestures 👋
git clone https://github.com/your-username/signlangai.git
cd signlangai
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
python manage.py migrate
python manage.py runserver