The Describatron

Turn lectures into study-ready guides.

What you'll get →

An accessible study guide rendered from your lecture video (or slide images) and a transcript. Each section contains:

  • A slide image — the most representative frame from that section's time range, with alternates you can switch to in the editor.
  • The on-screen text from the slide (Tesseract OCR by default; Claude OCR when an API key is provided).
  • The narration paragraph for that section, drawn from the transcript and stitched so sentences don't break across sections.
  • Optional AI enrichments when an API key is set — section titles, alt-text drafts, key terms, math equations, and topic-based grouping that produces a JamWorks-style chapter breakdown instead of one segment per slide.

The default output is a Review & Edit page in your browser where you can refine before downloading the final HTML or ZIP.

Generating your study guide…

starting Uploading…


      
      
    

Generate study guide

1. Upload

No video, just slide exports? Build from slide images + a transcript instead.  ·  Have a shared edits file? Open it to keep editing.

🔒 Used once for this run — sent over HTTPS, held in memory only, never written to disk or logs.

Tip: preview first so you only generate the slides you'll keep.

2. AI assist (optional)

With an API key above, Claude can enrich each segment. Toggle whichever you want; each runs only when checked.

Per-segment enrichments

Run once per kept slide. Cost scales with how many segments make it past the picker.

Whole-document features

Run once for the whole video. Change the document's overall structure.

3. Media extras

Local-only enrichments — no API key needed.

Processing usually takes ~30 s per minute of video.