EduBuddy

Inspiration

We were inspired by the "Education" challenge at the hackathon. We wanted to make something that we, and other students, could use in our lives.

What it does

EduBuddy is an application that takes whatever is in the user's clipboard as well as audio input and gives tools useful to students (quiz generation, summaries, math calculations, etc...).

How we built it

We built EduBuddy using ChatGPT4, Langchain, Hume, Whisper-1, pyttsx3 (text-to-speech), and Llama-Index.

Challenges we ran into

Challenges we ran into were primarily with sockets, particularly interacting with the front-end Tkinter app and our backend GPT/Langchain code.

Accomplishments that we're proud of

We're proud of eventually making our socket issues work and fixing our issues. We are also very proud of utilizing a lot of tools given to us, specifically ChatGPT4, Langchain, LLama-Index, Hume, and Whisper-1.

What we learned

We learned a lot about organization, planning, AI, and ML, as well as how to keep working under sleepless conditions.

What's next for EduBuddy

The next step for EduBuddy is to continue audio development. Currently, EduBuddy has functions for translating different languages in text into English, different languages in Audio to English, etc. As EduBuddy is cross-platform as well as potentially compatible with multiple languages, EduBuddy has the potential to become a useful student companion for students around the world.

Built With

• gpt4
• hume
• langchain
• llama-index
• python
• whisper-1