Inspiration
This idea was inspired by a real-life situation when we were attending a lecture our faculty who is required to cover the lecture within an hour rushed through the notes quickly leaving us with incomplete notes so one of our team members came up with an idea: what if we recorded these lectures and then summarized them for future revisions? This is how we came up with Gist Me.
What it Does
Gist Me is a cross-platform application that can summarize any text-driven PDFs into manageable, summarized versions.
How We Built It
We used OpenAI's GPT-3 API and utilized its two different APIs for summarization and bullet points to create concise and insightful PDFs.
Challenges We Ran Into
1) We tried to integrate Google Cloud's voice-to-text transcription to convert live lectures into summarized bullet points but a necessary package was not up to date causing codex issue. 2)We also implemented a custom audio recorder that takes audio input from the user and plays the recorded audio, but it wasn't as successful as we had hoped because we were unable to implement the transcription.
Accomplishments That We're Proud Of
We successfully implemented OpenAI's GPT-3 APIs, which summarize text efficiently.
What We Learned
This hackathon had many ups and downs, with some setbacks and some essential new learnings. Overall, this was the best hackathon experience we've ever had.
What's Next for Gist Me In future iterations of our application, "Gist Me," we plan to integrate Google Cloud's voice-to-text transcription to reach a wider market.
Comments