Inspiration
We wanted to do something with hand gestures and An ASL to Speech seemed it would be very useful for people with speech disability.
What it does
ASL gestures to Speech.
How we built it
Get images from opencv feed and feed it to mediapip and compare the resultes with the model upon which we play the audio for the gesture.
Challenges we ran into
Training a model is difficult with the limited and noisy data that we had
Accomplishments that we're proud of
Somewhat works
What we learned
- python
- mediapipe
- tensorflow
- keras
- opencv
- model training
What's next for ASLTranslator
- Fix the model
- move from images to mediapipe solutions
Log in or sign up for Devpost to join the conversation.