home page where user gets prompted to select a language
computer vision doing real time translation/definition
Sixty-two percent of English speakers wished they were better at speaking other languages and had plenty of reasons for doing so. Voga aims to minimize the language barrier for English speakers and foreign individuals. We developed an easy and accessible method to encourage people around the world to learn foreign languages and explore other cultures.
What it does
Voga provides a solution to those people who want to learn a new language but are intimidated by the language barrier. Voga allows people to use their device's camera to capture objects that they want to know in the foreign language. Simply point the camera to an object that you wish to know the foreign language of and Voga will save the word as a flashcard so the user could look back on later.
How we built it
The back end application is built with Python on the Flask framework hosted on AWS. We pipelined the data onto a NoSQL Firebase. We also hosted a Node.js server in order to communicate to the Flask server.
The computer vision ai is built with the COCO model on TensorFlow. We manipulated the boundary box data and specific object classifiers so the application would only perform prediction tasks on certain objects.
All of the artwork and UI was created with Adobe Illustrator.
Challenges we ran into
We had a lot of difficulties with MongoDB Atlas. We decided to go in a more simpler direction and use Firebase as our cloud storage. We also had a lot of trouble interacting between requests due to Cross-Origin Resource Sharing, however, we were able to find a way around it using a Node.js server.
Accomplishments that we're proud of
We were able to create a polished project in only 36 hours. The application is highly scalable since we use cloud API technology and we’re incredibly excited to create an application that could benefit the lives of many.
What we learned
We learned how to get around Cross-Origin Resource Sharing by communicating with the back end through a server that does not pass requests through a browser. We also learned how to connect to Firebase using the Firebase admin SDK python extension.
What's next for voga
We hope to make a standalone mobile app. We also hope to improve our A.I. because as time goes on, it will become outdated.