Skip to content

Latest commit

 

History

History
35 lines (24 loc) · 1.08 KB

readme.md

File metadata and controls

35 lines (24 loc) · 1.08 KB

Sarah for IISC Bengaluru Hackthon

The summarization tool first accepts input in three forms: topic title, user-provided content, or uploaded PDF. If a topic is given, relevant descriptions are scraped from Fox News, Fox Business, Fox Sports, and NDTV. Next, the data is fed into a fine-tuned BERT/Facebook summarizer model, followed by keyword extraction using spaCy. The tool then provides a summarized text along with keywords and links. For user-provided content or PDFs, the same process applies, followed by translation to Indian languages and optional voice reading.

landing page

Output images

by content

instructions

  • for backend
pip install -r requirements.txt
python3 app.py
  • for frontend
cd frontend
npm install
npm run dev
  • visit http://localhost:5173 on your pc