Project Outspoken

In lieu with the theme for Startup Weekend 2018, Social Good, we decided to build something for the hard of hearing community.

This is a web application that allows you to collect sign language images, train the model and finally deploy it in your web browser. All the user needs is a webcam. The application is able to detect sign language and translate it into captions.

After a weekend of pure hardcore hacking, we managed to get it to work and eventually won 2nd price at the hackathon!

Kudos to my team mates : @horensen, @gohjohn for making this possible!

Reference: This repo is originally a fork from TensorFlow.js Examples

TensorFlow.js.

Link to competition video

Here is the link to the pitch and demo!

https://www.dropbox.com/s/95yx7d03ywuangn/Outspoken%20pitch.mp4?dl=0

Dependencies

• Node.js version 8.9 or higher
• NPM cli OR Yarn

How to build an example

cd into the directory

If you are using yarn:

cd mnist-core
yarn
yarn watch

If you are using npm:

cd mnist-core
npm install
npm run watch

Details

The convention is that each example contains two scripts:

• yarn watch or npm run watch: starts a local development HTTP server which watches the filesystem for changes so you can edit the code (JS or HTML) and see changes when you refresh the page immediately.

• yarn build or npm run build: generates a dist/ folder which contains the build artifacts and can be used for deployment.

Contributing

If you want to contribute an example, please reach out to us on Github issues. We would greatly appreciate it if you can contribute datasets as we need highly diverse and massive amount of data in order to make this commercially ready.