Practice fingerspelling in American Sign Language (ASL) and Indian Sign Language (ISL) with live camera feedback.
Real-time hand recognition powered by MediaPipe. Free, no signup, runs in your browser.
SignLearn is a free, webcam-based tool for learning and practising sign language hand gestures. Pick a sign, see the target hand shape and a description of how to make it, then turn on your camera and try it yourself.
It is built by Mayank (Wondermayank) for ThunderStudy learners, and aims to make sign language practice more accessible to everyone.
- 🇺🇸 ASL and 🇮🇳 ISL modes: switch between American and Indian Sign Language
- 🔤 Alphabet: practise fingerspelling letter by letter
- 💬 Words & Phrases: a second tab for practising beyond single letters
- 🖐️ Target shape and description: each sign comes with a reference and a short how-to
- 📷 Live camera practice: start your webcam and compare your hand with the target
- 🎯 Real-time recognition (ASL): automatic gesture checking using MediaPipe hand tracking
- ✅ Progress tracking: mark signs as practised and see your "practised" count and percentage
- ⬅️➡️ Previous / Next navigation to move through the signs at your own pace
- 🆓 Free to use, no signup required
- ISL: Indian Sign Language fingerspelling mostly uses two hands to trace the shape of each letter, and signs can vary by region. For ISL, SignLearn shows reference descriptions and keeps camera tracking active for practice, but automatic correctness grading is currently available for ASL only.
- Practice mode is best-effort. SignLearn is a practice aid, not a certified sign language grader. For formal learning, please use a qualified teacher or an official course.
- Good lighting, a plain background and keeping your hand fully in frame give the best recognition.
Open https://project.wondermayank.in/fun/Sign_language, choose ASL or ISL, pick a letter and click Start camera & practice. Allow camera access when your browser asks.
The whole app is a single file, Sign_language.html.
# 1. Clone the repository
git clone https://github.com/wondermayank/Sign_language.git
# 2. Go into the folder
cd Sign_language
# 3. Serve it locally (the camera needs localhost or HTTPS)
python -m http.server 8000Then open http://localhost:8000/Sign_language.html.
Browsers only allow camera access on
localhostor HTTPS, so opening the file directly withfile://may not work. An internet connection is also needed to load the MediaPipe files.
Sign_language/
├── Sign_language.html # The complete app (UI, styles and logic)
└── README.md # Project documentation
| Layer | Technology |
|---|---|
| Markup | HTML5 |
| Styling | CSS3 |
| Logic | Vanilla JavaScript |
| Hand tracking | MediaPipe |
| Camera | Browser getUserMedia (WebRTC) |
- Automatic grading for ISL two-hand letters
- More words and phrases
- Saved progress across visits
Suggestions, corrections and new signs are welcome. Sign language accuracy matters, so please share a reliable source when suggesting a change.
- Fork this repository
- Create a branch:
git checkout -b feature/my-feature - Commit your changes:
git commit -m "Add my feature" - Push the branch:
git push origin feature/my-feature - Open a Pull Request
- 🌐 Website: wondermayank.in
- 📧 Email: contact@wondermayank.indevs.in
- 🐙 GitHub: @wondermayank
Free to use for learning and personal practice. Please give credit to Wondermayank / ThunderStudy if you share or reuse it.
Made with ❤️ in India by Wondermayank · Built for ThunderStudy learners