A private Wiki-style knowledge base generator from a collection of meeting audio transcripts.

What inspired us:

We noticed that many organizations struggle with organizing and retaining knowledge that's being shared during meetings. Much valuable information can be lost or forgotten even when notes are taken. That's why we wanted to create a tool that automatically extracts insights from meeting audio transcripts and generates a private knowledge base that team members can refer to.

How we built our project:

We first gathered our test data from podcast transcripts and developed our own in-house speech-to-text recognition app with AWS moving forward. With the help of GPT, we created a Desktop application that summarizes the content and uses Graphql to host a web-based Wiki server.

Challenges we faced:

It is our first time learning new technologies in a short period of time, for example, we created the apps that needs multi-threading

Logs:

  • 10:18 PM 3/11/23, the GPT summarizing and templating module is finished.
  • 3:27 AM 3/12/23, automated wiki creation with Graphql API call
  • 8:39 AM 3/12/23, created wiki documentation for an markdown folder

Built With

Share this project:

Updates

posted an update

Day 1 Recap (Chau): 11:00-ish PM: Began working on the GPT summarization component. 3:00-ish PM: Built GPT index, began testing queries for summarization. 10:18 PM: The GPT summarizing and templating module is finished. 12:14 AM: Ported over functions to a WikiGen class - cleaner code.

Log in or sign up for Devpost to join the conversation.