#project overview pip install -r requirements.txt

to run application: streamlit run main.py

Steps:

1. Extract Transcript
   • A. Directly from yt API
   • B. ASR(Automatic speech recognition)
     • i. Whisper OpenAI model (heavy)
2. Transcript summarization
   • A. Abstractive
     • i. Pegasus
   • B. Extractive
     • i. nltk
     • ii. Spacy
3. Metrics(To validate accuracy)
   • A. BertScore
   • B. Rogue
   • C. Meteor

Screenshot:

The Authors of this project are:

1. Sanjeevani Lakade
2. Shruti Gupta
3. Spruha Thorat
4. Tridib Nandi