ASL-Sign-Recognition is a real-time American Sign Language (ASL) recognition system. It helps users track 21 key hand landmarks using MediaPipe technology. This tool classifies gestures ranging from A to Z with a TensorFlow neural network. The application includes a dataset capture tool, training pipeline, and live webcam prediction. It works smoothly on any laptop without needing a GPU.
To get started, visit this page to download: ASL-Sign-Recognition Releases.
- Operating System: Windows, macOS, or Linux
- RAM: Minimum 4 GB
- Disk Space: At least 500 MB free
- Webcam: Required for live capture
- Click on the link above or visit the Releases page.
- Look for the latest version.
- Select the appropriate file for your operating system.
- Click the download link to start.
- Once the download completes, locate the downloaded file on your computer.
- Double-click the file to initiate the installation process.
- Follow the on-screen instructions to complete the installation.
- Make sure your webcam is connected.
After installation, you can start using ASL-Sign-Recognition.
- Open the application.
- Allow it to access your webcam.
- Position your hands in front of the camera.
- The application will recognize your ASL gestures in real-time.
- Real-time ASL gesture recognition.
- Utilizes MediaPipe for hand tracking.
- Pre-trained models for accurate recognition.
- Simple interface for easy navigation.
You can also use the dataset capture tool included in the application. This tool allows you to record your own gestures for later use in training.
- Open the dataset capture tool.
- Follow the prompts to record your hand gestures.
- Save your recordings for future training use.
If you encounter issues:
- Ensure your webcam is functioning properly.
- Check that you have the latest version of the application installed.
- Restart the application if it stops responding.
For additional support, please visit our issues page.
Join our community by following us on social media or GitHub discussions. Share your experience, ask for help, or collaborate with others interested in ASL recognition.
This project is licensed under the MIT License. Feel free to modify and use the software as long as you include original attribution.
For more information about ASL and its applications, you can visit various online resources and communities dedicated to sign language recognition.
- Download the app: ASL-Sign-Recognition Releases
- Report issues: GitHub Issues