Lingua - app for learning languages

The UI

Inspiration

Challenges of learning vocabulary for distinct scenarios. As one of our teammates is a non-native speaker of English, finding it easy to converse with friends but harder to learn situational conversations (doctors appointment, job interview, etc), we thought this could be a helpful tool for other people that may struggle.

What it does

It generates a conversation based on a scenario provided by the player, where the player assumes the role of a speaker conversing with another speaker in the given context. The player has to fill in missing words for the responses of one of the characters, scoring points for each valid response and getting feedback if they went wrong.

How we built it

Using google bard as a chatbot, we generated conversations and graded responses, with the backend implemented in python. On the frontend, we used typescript to display the conversation between the chatbot and the player. Eleven labs was used to output audio

Challenges we ran into

Getting the chatbot to return appropriate responses to our prompts, connecting the frontend to the backend.

Accomplishments that we're proud of

Managed to connect backend and frontend. Dynamic rendering. Text to speech.

What we learned

Learned how to properly set up prompts, how to use typescript in react, how to use eleven labs and to implement text to speech, working in a team for development.