DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to turn a lecture video into a synced notes page with Lecture to Notes

Transcribe locally, extract slides, ground them to speech, and export one HTML page with video, transcript and summary.

  • Research
  • Advanced
  • 2 hours
  • From the README
  • Any agent

Who it is forA clinician who recorded a hands-on course and wants to find specific demonstrations later.

The problemNobody rewatches hours of video to find one moment.

The outcomeA viewer where clicking a note timestamp seeks the video to that moment.

What you need first

  • Python with the requirements files
  • A GPU for local Whisper transcription
  • Ollama with minicpm-v:8b for optional slide semantics
  • An LLM for the final synthesis stage

Steps

  1. Install requirements and pull ollama pull minicpm-v:8b for slide semantics
  2. Transcribe with transcribe_video.py, passing --lang explicitly
  3. Run the slide extraction, OCR and grounding stages printed by route_inputs.py
  4. Export the synced HTML viewer with export_web.py

Prompt for your agent

Process <lecture.mp4> with drpwchen/lecture-to-notes, passing the spoken language to --lang, and export the synced HTML viewer. Verify three random note timestamps seek to matching transcript lines.

Caveat. Transcription has no default language on purpose; a wrong guess makes Whisper produce fluent but wrong text.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for drpwchen/lecture-to-notes and "Turn a lecture video into a synced notes page"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer