Linguist.AI

Inspiration

We wanted to help people connect across cultures and languages, especially those who face language barriers or different hearing abilities. Many translation tools focus on output over understanding. Linguist.AI was inspired by the idea that everyday communication can also be a learning opportunity, making conversations more inclusive while helping users actually learn a new language.

What it does

Linguist.AI is a real-time translation and language learning platform. Users can speak in one language and instantly receive translations in their preferred language. Unlike traditional translators, Linguist.AI explains why a translation works by breaking down grammar, vocabulary, and context based on the user’s skill level.

Users can listen to both the original and translated audio to practice repetition and pronunciation. A daily learning dashboard summarizes conversations, highlights grammar patterns, and introduces new vocabulary, turning real conversations into structured learning.

How we built it

We built the frontend using Next.js with React and TypeScript to create a responsive, user-friendly interface. The backend relies on Python Flask and Supabase (PostgreSQL DB, Auth) for secure data management and user authentication. AI-powered translation and personalized learning explanations are delivered through Gemini via OpenRouter, while ElevenLabs provides realistic AI voice synthesis for clear, natural speech. Tech stack:

Frontend: Next.js with React and TypeScript for a smooth, responsive interface
Backend: Flask + Supabase (PostgreSQL DB, Auth) for secure data storage and user management
AI/LLMs: Gemini via OpenRouter for LLM-powered translation and personalized learning explanations, ElevenLabs for realistic AI voice synthesis

Challenges we ran into

Translating audio in real-time and then playing it back smoothly was a challenge, and we spent a lot of time experimenting to find the right AI models that had accuracy and matched our use case. Working with API limits added another layer of challenge, and designing the learning dashboard pushed us to think carefully about how to organize the database schema, what features to include, and how to process LLM responses effectively.

Accomplishments that we're proud of

We’re proud of taking Linguist.AI from an idea to a working MVP. The UI is polished and intuitive, and the core experience of translating while learning feels cohesive. Seeing everything work together showed us the real potential impact of the project.

What we learned

We learned that building real-time AI translation is as much about usability and accessibility as it is about the tech. Choosing the right AI model, organizing the data, and designing a dashboard that actually helps learners took experimentation and iteration. Most importantly, we saw how small design choices can make a big difference in helping people communicate and learn effectively.

What's next for Linguist.AI

We plan to expand support for multiple languages and dialects and add real-time pronunciation guidance. For learning mode, we endeavor to have adaptive dashboards that evolve with each user’s fluency. We also aim to enhance accessibility features to better support users with different hearing abilities.