LangLearn

• 

  This is the home page where users can track their learning progress.

• 

  Users can add and save sentences they want to remember. AI analyzes each sentence for correctness.

• 

  Users can choose from three review types to practice saved sentences. AI evaluates pronunciation and sentence fluency through voice input.

• 

  Users can view all saved sentences and take quizzes to test themselves.

• 

  This page shows a visual chart of the user's overall learning status.

Inspiration

While studying English through dramas and YouTube, I often jotted down useful sentences I wanted to remember—using Notion or sticky notes. But since I’m not good at organizing, I struggled to find those notes later. Another issue was that idioms and slang were often hard to look up in a dictionary, and interpreting them one by one with ChatGPT was time-consuming. That’s why I built this platform: to store and review sentences in one place, and to get AI-powered interpretations of idioms and slang.

What it does

Users can input sentences they want to remember, and AI extracts and explains any idioms or slang. The sentences are saved for later practice. For review, users are encouraged to speak aloud, and the AI evaluates pronunciation and sentence accuracy to provide feedback for improvement.

How we built it

The entire service was built using Bolt, relying heavily on prompt-based development. Except for setting up Supabase and OpenAI API keys, everything else was implemented through carefully crafted prompts.

Challenges we ran into

Functionally, most parts were accomplished using prompts. However, when it came to styling, prompts still fell short. Even though I fed Bolt with design references from Pinterest, the results lacked accessibility and responsive behavior. It felt like styling still required human intervention.

Accomplishments that we're proud of

We’re proud that the AI can evaluate users’ pronunciation and sentence fluency. This allows users to recognize and fill their gaps more effectively, making their speaking practice more targeted.

What we learned

We learned the importance of prompt engineering and the convenience AI can offer. Writing detailed and modular prompts is key to building functional services with AI.

What's next for LangLearn

We learned the importance of prompt engineering and the convenience AI can offer. Writing detailed and modular prompts is key to building functional services with AI.

Built With

• openai
• react.js
• supabase
• tailwindcss
• typescript