lecture-stt

Transcribe audio lectures into structured markdown notes with PDF slide context.

5|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/lidge-jun/cli-jaw-skills --skill lecture-stt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lecture-stt
Source: https://github.com/lidge-jun/cli-jaw-skills/tree/main/lecture-stt
Command: npx skills add https://github.com/lidge-jun/cli-jaw-skills --skill lecture-stt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, mlx-whisper, openai, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the transcription of audio lectures into structured markdown notes, making it easier to review and study complex material.

Core Features & Use Cases

  • LLM-based STT: Leverages advanced models like Gemini for high-quality transcription.
  • Contextual Prompting: Uses PDF slides and domain terms to improve accuracy and structure.
  • Multi-Engine Support: Falls back to local Whisper or OpenAI for flexibility.
  • Use Case: Convert a 2-hour university lecture recording into detailed, page-by-page markdown notes with mathematical formulas correctly formatted.

Quick Start

Use the lecture-stt skill to transcribe the audio file 'lecture.m4a' with the provided PDF slides 'slides.pdf'.

Frequently Asked Questions about lecture-stt

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

FAQPage Schema
How do I transcribe lecture audio into structured markdown notes?▼

This Skill transcribes lecture audio into structured markdown notes using LLM-based Speech-to-Text with contextual prompting. It processes audio recordings to output organized, page-by-page text by leveraging provided context.

Can I use PDF slides to improve speech-to-text accuracy for lectures?▼

Yes, you can provide PDF slide guides and domain-specific terms to improve transcription accuracy. The LLM uses this contextual prompting to correctly format mathematical formulas and structure the generated lecture notes.

Does this transcription tool support local Whisper as a fallback engine?▼

Yes, the tool supports local Whisper as a fallback engine for speech-to-text transcription. It primarily leverages Gemini and OpenAI models but switches to local processing to maintain flexibility and accurate output.

What is the best way to convert a 2-hour university lecture recording into detailed notes?▼

The best way to convert a 2-hour university lecture recording into detailed notes is using an LLM-based transcription tool. It transforms long audio files into structured markdown content with correctly formatted mathematical formulas.

Do I need an internet connection to use this lecture transcription Skill?▼

An internet connection is required for LLM-based transcription using Gemini or OpenAI. However, you can transcribe audio offline by falling back to the local Whisper engine for generating your markdown lecture notes.