Inspiration
Language acquisition is difficult - especially for transfer students, new immigrants, and tourists, sudden immersion in a foreign tongue can be overwhelming. We wanted to create a quick and easy way for language learners to create and collect flashcards from everyday life - and share them with similarly-minded friends too!
What it does
With LinguaML, you can take a picture of anything - we mean anything - and get it converted into a flashcard containing the subject's name in your native language and your target language, as well as the picture you took. You can hear the words spoken out loud, share your cards with your friends via QR code, and even convert entire decks into other languages, in case your focus shifts!
How we built it
We created a React Native app and used the Google Vision API to do image tagging. We used Microsoft Translation on Azure to translate the words. All other storage is done locally.
Challenges we ran into
Choosing how to initialize the project was difficult since each had differing compatibilities with certain frameworks and plugins. Moreover, connecting the React Native camera to the Google Vision API was incredibly difficult, as none of us had ever used GCP before, and half of us were first-time React Native users. Lastly, translation took a huge bulk of our time, but since it was hard to make that a group task, it gave other members much more time to polish other parts of the project up.
Accomplishments that we're proud of
We're proud that we were able to create what we thought was an incredibly complete and polished product. We not only hit all of the core functionality goals that we set out to achieve initially, but we also nailed every single item on our wishlist, and then some. Moreover, we're proud of the sheer amount we were able to learn from each other, and the fun we had along the way :).
What we learned
We all got a chance to brush up our React Native skills, as well as some more experience using third-party APIs and more obscure packages.
What's next for LinguaML
User authentication, publishing to the app store, and better image tagging technologies are all possibilities for LinguaML to grow to the next level.
Log in or sign up for Devpost to join the conversation.