This folder contains the model code for the app "Look & Tell"
- A deep-learning neural network for hand action recognition, made by Vietnamese students, to translate Vietnamese sign language
- The story begins when Kiet becomes my neighbor, he is a born deaf, ie since he was born, he has not been able to hear his mother's voice. And every day, for 10 years - until now, that child is in 5th grade, I have witnessed every day that the child struggles, has difficulty in expressing the simplest wishes of an elementary school child.
- Learn about the deaf and hard of hearing, there are currently 1.5 billion speech and hearing impaired people in the world, and 2.5 million people in Vietnam - a huge community! However, it entails a huge economic burden for countries. According to a 2015 WHO report, the cost of speech and hearing impairment ranges from $750 to $790 billion. By 2019 it will exceed $981 billion (according to Global Burden Disease), mainly due to community and social costs - an too great economic burden for countries. * In Vietnam, the number of qualified sign language interpreters is only about 20 people, and there are no reasonable communication support products (only 1 product is produced by a student of Ho Chi Minh City University of Science and Technology but requires an additional one). equipment costs nearly 1 million VND), while conducting a survey of 250 people including deaf and hard of hearing people, it was found that nearly 100% of people want to create a bridge to integrate and really need a product that supports communication.
- With the desire to talk with Kiet (my neighbor) as well as all the deaf and hard of hearing people in the world, my friend and I decided to do this project.
git clone https://github.com/khooinguyeen/Sign-Language-Translator.git
!pip install tensorflow==2.5.0 tensorflow-gpu==2.5.0 opencv-python mediapipe sklearn matplotlib
(3) go to the file RunModel.ipynb to play with ipython notebooks and run the model
- Relative path:
Sign Language Translator\RunModel.ipynb
(4) to collect more data by your own, go to the file CollectData.ipynb
- Relative path:
Sign Language Translator\CollectData.ipynb
- If you appreciate the humanity of the project or have a problem or even just want to talk to us, please contact us via email: [email protected]