lecture

Transcribe local lecture videos into vault-formatted notes with embedded slide screenshots.

36|2|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/tuan3w/obsidian-vault-agent --skill lecture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lecture
Source: https://github.com/tuan3w/obsidian-vault-agent/tree/main/skills/lecture
Command: npx skills add https://github.com/tuan3w/obsidian-vault-agent --skill lecture

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ffmpeg, ffprobe, mlx-whisper, python, and includes scripts (resource) components.

What problem does it solve?

This Skill removes the manual effort of transcribing a local lecture video and turning it into readable, vault-formatted notes, so you can study the material without rewatching the recording.

Core Features & Use Cases

  • Transcript extraction (multi-language): Transcribes local video audio using mlx-whisper and produces an English note, auto-detecting the transcript language.
  • Slide screenshot capture: Extracts key frame screenshots with timestamps from the video and embeds them into the final Obsidian note.
  • Vault-structured synthesis: Organizes the lecture into thematic sections using a dedicated lecture-noter agent, and searches for related existing notes to connect ideas.
  • Use case: When you have a course recording or class video on disk, this generates a study-ready lecture note with embedded slide evidence and cross-links.

Quick Start

Use the lecture skill on your local file by running: /lecture /path/to/video.mp4

Frequently Asked Questions about lecture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I transcribe a local lecture video into Obsidian notes?

To transcribe a local lecture video into Obsidian notes, run the /lecture command on your MP4, MOV, MKV, AVI, or WEBM file. It extracts audio with mlx-whisper and generates vault-formatted notes with embedded slide screenshots.

Can I extract slide screenshots with timestamps from a local video recording?

Yes, you can extract slide screenshots with timestamps from a local video recording. The skill uses ffmpeg to capture key frames and embeds them directly into the synthesized lecture notes for visual reference.

Does mlx-whisper support multi-language video transcription for offline recordings?

mlx-whisper supports multi-language video transcription for offline recordings by auto-detecting the spoken language. It processes the extracted audio and produces an English note for your Obsidian vault.

What video formats are supported for generating structured lecture notes?

Supported video formats for generating structured lecture notes include MP4, MOV, MKV, AVI, and WEBM. The skill validates the input path and video type before extracting audio and capturing frames.

Do I need ffmpeg and python installed to transcribe local class videos?

Yes, you need ffmpeg, ffprobe, python, and mlx-whisper installed to transcribe local class videos. These dependencies handle audio extraction, frame capture, and speech recognition for the final note synthesis.

How does cross-linking work for newly generated lecture notes?

Cross-linking for newly generated lecture notes works by optionally searching your existing Obsidian vault for related notes. The skill connects thematic sections from the current lecture to relevant existing ideas.