The Project writeup is in the zip file-- please download it to view
Inspiration
Our team member, Sarah, is from Malaysia, where pidgin languages are common. Growing up speaking primarily a pidgin of Malay and English as well as casual, informal Malay, she does not speak any formal Malay. Malay is not provided on Duolingo and she realized the opportunity to build a language software that supports more languages. With the Speak plug-in, the prompted AI was able to not only speak and teach in formal Malay, the AI was able to speak in Malay from different communities (in a Muslim community and pidgin Malay-English). To our surprise, the pidgin was modern, relevant and also grammatically correct with the societally defined language.
What it does
The program has the ability to converse in a language of choice, providing translations to AI input and responding to the user. It has the ability to understand slang and teach formal language instead. The app has been prompted with restrictions on how to act and to let the user know when they have made a mistake. We believe there's so much potential with this idea to help people learn new languages and translate between languages that are often overlooked.
How we built it
We built the app using a Swift tech stack on Xcode with a Storyboard front end. We also used the OpenAI Chat API. More OpenAI API's could be used in the future to further expand the app.
Challenges we ran into
Remembering how to code in Swift and handle storyboard was something that was very difficult and overall, ideating was a long process for us.
Accomplishments that we're proud of
We're proud of coming up with an idea that is impactful and briefly implementing it to display what is needed.
What's next for Language Buddy
Expanding the apps' UI, it's functionality, having more user-friendly error checks, and using more API's to expand its features (text-to-speech is definitely up next!).
Log in or sign up for Devpost to join the conversation.