Medley

Medley is a tiny library that lets you parse documents of different formats like pdf, pptx, html or wav.

If you are not sure if this library right for your task, you might want to see how I use it

• Parse documents of different formats for my RAG applications
• Transcribe audio files (supports around 100 languages)
• Identify and sanitize PII information for upstream services

Supports

• Text [.pdf, .pptx]
• Audio [.wav]

WIP

• Text [.html]