PaperParser transforms academic research papers into podcasts π§ and PowerPoint presentations π.
By combining AI summarization and automated content generation, it makes scholarly content easier to consume and share.
- π§ Podcast Generation β Listen to research papers in audio format
- π Presentation Creation β Auto-generate slide decks with key insights
- π PDF Parsing β Extract text and structure from uploaded research papers
- βοΈ Cloud Integration β Secure storage and authentication with Supabase
- β‘ Node.js + Express.js β Core API server
- π§ Gemini API β Summarization and content generation
- π PyMuPDF β High-quality PDF text parsing
- βοΈ Supabase β Database + authentication layer
- π³ Docker β Containerized backend for easy deployment
- βοΈ Next.js (React + TypeScript) β Modern frontend framework
- π Tailwind CSS β Utility-first styling for responsive UI
git clone https://github.com/P1Manav/PaperParser.git
cd PaperParserNavigate to backend:
cd backendCreate a .env file (sample):
SUPABASE_URL=
SUPABASE_SERVICE_ROLE_KEY=
GEMINI_API_KEY=
google_api_key=
PORT=5000
NODE_ENV=Build Docker image:
docker build -t paperparser_image .Run the container:
docker run --env-file .env -p 5000:5000 --name paperparser_container paperparser_imageπ Backend runs at: http://localhost:5000
Navigate to frontend:
cd ../frontendCreate .env.local file (sample):
NEXT_PUBLIC_SUPABASE_URL=
NEXT_PUBLIC_SUPABASE_ANON_KEY=
GEMINI_API_KEY=
NEXT_PUBLIC_API_URL=http://localhost:5000Install dependencies:
npx yarn installRun development server:
npx yarn devπ Frontend runs at: http://localhost:3000
- π Upload a PDF in the frontend
- βοΈ Choose output β Podcast π§ or Presentation π
- β¬οΈ Download or play the result directly
Contributions are welcome! π
- Open an issue for bugs or feature requests
- Submit a PR to help improve PaperParser