Reader Waypoint

Listen to any PDF

Pick up where you left off. Nothing leaves your device.

Drop a PDF here, or .

  • Private by design. Reading, speech and OCR all run in your browser. Your file is never uploaded.
  • Natural voices. Use your system voices, or download an on-device neural voice once.
  • Scans too. Pages without selectable text can be read with on-device OCR.

How your privacy is protected

Credits and open-source licences

This site is built on free, open-source work. Thank you to the people behind it:

  • Kokoro-82M voice model by hexgrad (Apache-2.0), run with kokoro-js and Transformers.js from Hugging Face (Apache-2.0) and ONNX Runtime Web from Microsoft (MIT).
  • PDF.js from Mozilla reads PDF files (Apache-2.0).
  • Tesseract OCR, via Tesseract.js (Apache-2.0), with English and German language data (MIT/Apache-2.0).
  • Atkinson Hyperlegible typeface from the Braille Institute (SIL Open Font License 1.1).
  • idb by Jake Archibald for browser storage (ISC).

Download the natural voice?

This fetches a speech model (about 90 MB) from Hugging Face once and keeps it in your browser, so it works offline afterwards. Speech is then generated on your device. Your documents are never sent anywhere.

Appearance and settings

Theme
Reading font