Hack Jaipur

Team Name :

Runtime Terror

Name of Project

Multi Office Summarizer

Tagline

An abstractive summarizer for office needs

The problem it solves

It is used to summarise meetings by converting speech to text. The text file is then sent to the summariser to generate a summary. We have also built an optical charcter recognition system so that we can extract text data from pdf and convert it to a txt file and this txt file will also be fed to the summarizer.

Challenges we ran into

Running the Tesseract and other related libraies we ran into dependency issues and finally we overcame it by using Linux and was not working with Windows.

Technologies we used

Python , HTML5 , CSS3 , Tensorflow , NLTK , PIL , Pillow , Pandas , Numpy , Tesseract