README.md

metadata

title: PaperCast
emoji: 🎧
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 6.22.0
app_file: app.py
pinned: false
python_version: '3.11'

Papercast

Papercast converts scientific papers into structured summaries and listenable audio explanations using multimodal document parsing, scientific summarization, and neural text-to-speech.

Current features

• PDF and arXiv paper ingestion
• Standard Docling and Granite-Docling document processing
• Extractive and PEGASUS-X summarization
• Technical narration preparation
• Neural text-to-speech output

Limitations

• Large transformer models may be slow on CPU hardware
• Equation narration is currently approximate
• Complex papers may require processing in smaller page ranges