# AI use cases for Lecture to Notes (drpwchen/lecture-to-notes)

Repo: [drpwchen/lecture-to-notes](https://github.com/drpwchen/lecture-to-notes) | Use cases: 3 | Page: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/

lecture-to-notes is a local GPU pipeline and Claude Code skill that turns lecture recordings into slide-illustrated notes and a synced HTML viewer where video, timestamped transcript and summary sit on one page.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Get the exact processing plan for a folder of lecture files: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/plan-lecture-processing-from-folder/
- Turn a lecture video into a synced notes page: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/turn-lecture-video-into-synced-viewer/
- Let Claude Code run the whole lecture pipeline: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/let-claude-code-run-lecture-pipeline/

## How to get the exact processing plan for a folder of lecture files with Lecture to Notes

Repo: [drpwchen/lecture-to-notes](https://github.com/drpwchen/lecture-to-notes) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README

Drop video, audio, PDF decks or slide images in a folder and get the ordered commands to run.

**Who it is for:** A student with a recorded course, a slide PDF and no idea where to start.

**The problem:** Different input mixes need different pipelines.

**The outcome:** A printed, plan-only list of commands plus the questions a human must answer first.

### Steps

1. Clone with `git clone https://github.com/drpwchen/lecture-to-notes` and install requirements
2. Copy the config with `cp config.example.yaml config.yaml`
3. Run `python scripts/route_inputs.py /path/to/material_folder`
4. Answer the listed questions, then run the printed commands

### Prompt to paste into your agent

```text
Run `python scripts/route_inputs.py <folder>` from drpwchen/lecture-to-notes with --json and list each step with whether it needs GPU or network. Confirm nothing was written to disk.
```

### Caveat

route_inputs.py never runs anything; you or your agent still execute each command.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/plan-lecture-processing-from-folder/

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

Repo: [drpwchen/lecture-to-notes](https://github.com/drpwchen/lecture-to-notes) | Difficulty: Advanced | Time: 2 hours | Basis: From the README

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

**Who it is for:** A clinician who recorded a hands-on course and wants to find specific demonstrations later.

**The problem:** Nobody rewatches hours of video to find one moment.

**The outcome:** A viewer where clicking a note timestamp seeks the video to that moment.

### 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 to paste into your agent

```text
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.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/turn-lecture-video-into-synced-viewer/

## How to let Claude Code run the whole lecture pipeline with Lecture to Notes

Repo: [drpwchen/lecture-to-notes](https://github.com/drpwchen/lecture-to-notes) | Difficulty: Advanced | Time: 2 hours | Basis: From the README

Install the repo as a Claude Code skill so the agent drives every stage and writes the final notes.

**Who it is for:** A researcher who wants polished notes from conference talks without babysitting scripts.

**The problem:** Running a dozen pipeline stages by hand is tedious.

**The outcome:** Agent-driven notes where every claim links to a transcript moment and slide.

### Steps

1. Copy the repo tree into ~/.claude/skills/lecture-to-notes/
2. Set up requirements and config.yaml
3. Ask Claude Code to process a material folder
4. Review the audit output for unsupported claims

### Prompt to paste into your agent

```text
Using the lecture-to-notes skill, process <folder> into notes and the HTML viewer. Report any claims audit_note.py flagged and confirm each summary point cites a timestamp.
```

### Caveat

Stages 0 to E run standalone, but the final prose synthesis needs an LLM.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/drpwchen-lecture-to-notes/let-claude-code-run-lecture-pipeline/