FeedChevron right imageMobile DevelopmentChevron right imageSignBeeChevron right image
Post main image


Duolingo for Sign Language alphabet.

SignBee is a game designed to make learning fun. Our app is a first step in the right direction to help people get familiarized with American sign language--starting with the alphabet! We chose to create a Progressive Web Application to make SignBee both web browser and mobile-friendly. We built our app in Node.js using React, and used Firebase for all of our database and hosting needs. We used Figma to conceptualize our app and design high fidelity wireframes. In order to style our application, we took advantage of Material-UI and CSS. To make accurate gesture detections, we used Tensorflow.js which is an open-source machine learning library to design, build and train deep learning models. For SignBee, we took advantage of Tensorflow.js’s pre-trained Handpose model that allows real-time hand detections. We also added a pre-trained Fingerpose model to allow real-time detections of finger positions. This allowed for custom gestures to be used in our game. For our backend, we used Firebase as it provides an all-in-one solution that met all our app’s needs. The Cloud Firestore NoSQL database made sense because it allows for faster read times compared to relational databases. The data is updated in real-time and is accessed with simple queries. User authentication, static file storage, and deployment were also a part of the Firebase suite and very intuitive to set up and manage. Team Members: Goncagul "Gloria" Ay, Julia Kravets, Leon Zhao and Naomi Diaz

Video demo
Project screenshot 1
Project screenshot 2
Project screenshot 3
Comment post

Skip straight to final coding interviews with
Remote Companies!

Top Makers

Most active posts

Top projects