/Gist

More than just minutes of the meeting.

Primary LanguagePythonMIT LicenseMIT


More than just Minutes of the Meeting

Gist

More than just minutes of the meeting! 🚀

Motivation

Both of us in the team are getting back to college after a considerable amount of time spent as professionals. One common thing we noticed was how expensive meetings really are. Say, an engineer costs about 100 USD an hour. A meeting of 10 such engineers for an hour would amount to $1000. Keeping this in mind, it becomes important to make meetings more efficient.

If conducted efficiently, meetings are an essential part of any organizations and helps teams collaborate on multiple projects.

However, meetings inherently have a frequent problem: Not all required attendees are able to attend the meeting at all times. This provides a clear gap where those who could not attend it or need a gist of what the meeting was about, what the action items were at the end of it and what the most important points discussed were, are left clueless and have no way of acquiring this information.

How to use Gist

  • Upload meeting transcript using the option in the sidebar.
  • Click on Generate Insights to see the most important insights from the meeting.
  • Click on Generate Summary to get a TL;DR of the meeting.

Domain

The domain for this project is : Text Summarization with focus on summization for meeting transcripts.

According to Tom Mitchel, “A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.”

For summarization on Amazon Food Review Dataset

=> P(T, E + ΔE) > P(T, E)

  • Performance(P): Sparse categorical loss while training the model.
  • Task(T): Summarization of food reviews.
  • Experience(E): Textual review data from Amzon Food Review Dataset.

Code Requirements 🦄

You can install Conda for python which resolves all the dependencies for machine learning.

pip install -r requirements.txt

Execution 🐉

streamlit run st_analyzer.py

Results 📊

@misc{vaswani_shazeer_parmar_uszkoreit_jones_gomez_kaiser_polosukhin_2017, 
   title={Attention is all you need}, 
   url={https://arxiv.org/abs/1706.03762}, journal={arXiv.org}, 
   author={Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N. and Kaiser, Lukasz and Polosukhin, Illia}, 
   year={2017}, 
   month={Dec}
 } 

References 🔱

Made with ❤️ and 🦙 by Abhinaav Singh and Akshay Bahadur