Inspiration
Wikipedia is an app that provides a wealth of information today. We wondered if it would be possible to transport its many articles through time to gain a comprehensive view of the future and past.
What it does
Chronopedia takes Wikipedia articles and procedurally updates them based on the year and language provided by the user. Text-to-speech functionality is also provided, making it more accessible.
How we built it
We employed a Flask backend to grab the Wikipedia files, using DeepSeek LLM and the Python Huggingface library to modify the Wikipedia file text and send it to the Svelte frontend. The frontend sends out a request every time the search bar, year, or language values are updated.
Challenges we ran into
Originally, when we used VSCode Liveshare to collaborate, the backend threw a connection error every time it got a request. Eventually, we solved the problem by moving all files to a single computer and running the program on Windows OS instead of macOS.
Accomplishments that we're proud of
We're proud that the Deepseek AI LLM was able to successfully generate realistic Wikipedia articles from the future and the past, properly updating tenses and years
What we learned
We learned that using VSCode to collaborate when using Python and Svelte.js is not a good idea when members of the team are unfamiliar with the IDE.
What's next for Chronopedia
Next, we plan to integrate image generation into the Wikipedia articles, making them more realistic and engaging.
Log in or sign up for Devpost to join the conversation.