robertsisaacp / community_board_meeting Goto Github PK

Making local policy information accessible and byte-sized

A public written record of community board meeting transcript and summary.

Explore transcripts »

Medium · Twitter · Request Feature

• About the Project
  • Built With
• Getting Started
  • Prerequisites
  • Installation
• Usage
• License
• Contact
• Acknowledgements

Community Board Meeting Minutes - NLP & Civic Tech

block party makes local policy information accessible and byte-sized, with a weekly email of your community board’s most recent meeting synopsis and transcript data.

block party generates summarizations of community board meetings directly from transcript data collected from YouTube.

Here's How:

• Collect closed captions from YouTube recordings of community board meetings
• Clean text and add grammar
• Extract key content and automatically generate summary

1. Pull Transcript from Youtube

   https://pypi.org/project/youtube-transcript-api/

2. Add Punctuation and grammar

   https://pypi.org/project/punctuator/

   Punctuator Model

   Ottokar Tilk and Tanel Alum{"a}e}, Bidirectional Recurrent Neural Network with Attention Mechanism for Punctuation Restoration, Interspeech 2016

3. Extract key content

   Honnibal, M., & Montani, I. (2017). spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing.

4. Summarize Text https://radimrehurek.com/gensim/summarization/summariser.html

Distributed under the GPL-3.0 License. See LICENSE for more information.

Sarah Sachs: @sarah_june42

Brandon Pachucua

Project Link: https://github.com/sarahJune1/community_board_meeting