AI teacher

Inspiration

Many people want to learn a new language but prefer not to let others know about it; even if they do start learning—using apps like Duolingo, for instance—they often keep it hidden due to the fear of judgment. The issue is that they usually spend only 5 to 20 minutes on these apps, and the learning stops the moment they close the application; they fail to integrate language practice into their everyday conversations

What it does

My idea is to create an AI app that facilitates conversation in any setting—whether in public, at the office, or elsewhere. Since almost everyone wears earbuds these days, the concept involves an AI language learning app that remains active (with the user's permission) even after the interface is closed. It would converse with the user via earbuds, correcting their mistakes in real-time and teaching the language. This allows the user to practice during their spare time without anyone else realizing it; to an observer, it would simply look like they are talking on a phone call.

How we built it

For demo use Flutter for Mobile app and for Android background service use Android Foreground Service and for Backend i use FastAPI, LLM API, Speech-to-Text API, Android TTS / cloud TTS for text to speech, Database with PostgreSQL, for Authentication use Firebase Auth, and Conversation history Firestore

Challenges we ran into

1. user trust

Accomplishments that we're proud of

What we learned

What's next for AI teacher

Built With

• android-tts
• fastapi
• firebase
• firestore
• flutter
• llm
• other
• postgresql
• speech-to-text