AI-generated executive team meeting minutes with OpenAI, Loops, Code, Google Docs, and Slack

  • 19 January 2023
  • 3 replies
  • 432 views

Userlevel 1

Hi, all!

I work with Chiefs-of-Staff on teams at VC-backed startups.

My clients find creating and sharing meeting minutes can be super time-consuming, so I created a Zapier pipeline that integrates automatically generated Google Meet transcripts with OpenAI to produce comprehensive meeting minutes, including:

  • A summary,
  • Key insights,
  • Top quotes,
  • Requests for help from the rest of the company, and
  • Feedback for the meeting facilitators/participants

This is then posted to the company Slack.

Here’s a tutorial on how to build a Zapier Zap that connect with your own data and automatically creates meeting summaries!

There’s so much more I’d like to do with this, including linking directly to video clips of the quotes, storing the transcript in a vector DB like Pinecone so it can be queried discursively later (many via a Slack bot w/ access to the company’s private knowledge repo?), and incorporating context in re the company’s quarterly goals into the prompts…

I had a blast creating this pipeline with Zapier’s new OpenAI support (and some of the other beta tools like Loops) and am looking forward to sharing more.


This post has been closed for comments. Please create a new post if you need help or have a question about this topic.

3 replies

Userlevel 1

@culstrup just letting you know this was very helpful for me!

I didn’t end up using this exact process, but I did end up utilizing your loop for creating segments of larger bits of text in multiple different zaps with Openai!

very cool, thanks so much for sharing.

Userlevel 1

@danbarrett316 glad to hear it! I’m thinking about building an API endpoint that handles the Chungking and pre-processing of longform text so it’s a lot easier to send to the open AI API

 

Is that something that you would find useful?

Userlevel 1

@danbarrett316 glad to hear it! I’m thinking about building an API endpoint that handles the Chungking and pre-processing of longform text so it’s a lot easier to send to the open AI API

 

Is that something that you would find useful?

Maybe! I don’t have an immediate need, but am assuming that it’s going to be an issue that repeatedly comes up.