hrescak

hrescak / transcribe-md

Public

Transcribe microphone and computer audio to markdown.

82
5
100% credibility
Found Mar 27, 2026 at 82 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A macOS tool that records your microphone and system audio during calls, transcribes speech locally into timestamped Markdown files distinguishing speakers.

How It Works

1
🔍 Discover easy meeting notes

While chatting about coding tools, you learn about a handy Mac helper that turns spoken meetings into neat written notes.

2
📥 Install with one command

Copy and paste a simple line into your terminal, and it sets everything up automatically on your Mac.

3
🔒 Grant listening permission

Tell your Mac it's okay for the terminal to hear your microphone and call sounds by checking a privacy setting.

4
🎤 Kick off transcription

In your coding assistant, type a quick note like 'start transcribing this meeting' and it begins capturing voices right away.

5
🗣️ Have your conversation

Speak freely during the call; it listens to you separately from the other person's voice and skips any repeats.

📝 Enjoy your ready notes

Stop recording, and instantly see a clean, time-stamped Markdown file labeling who said what, perfect for reviewing.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 82 to 82 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is transcribe-md?

transcribe-md is a Python tool for live transcribing microphone and computer audio to timestamped Markdown files on macOS 14+. It captures your mic as "You" and system audio (like call participants) as "Them," with automatic deduplication of speaker echoes, powered by local whisper.cpp with Metal GPU acceleration—no cloud services needed. Slash commands like `/transcribe-md meeting.md` kick it off right in Claude Code or Cursor, solving the hassle of manual note-taking during calls or brainstorming.

Why is it gaining traction?

Unlike cloud-heavy options like Amazon Transcribe or Google alternatives on GitHub, it runs entirely offline for privacy and speed, dumping clean, speaker-labeled Markdown in real-time without virtual audio hacks. The one-line installer handles ffmpeg, models, and IDE skills, plus options like `--mic-only` or `--duration 30` make it dead simple for quick sessions. Developers dig the GitHub transcribe Whisper integration that just works on Apple Silicon, beating buzzier Buzz or Vibe Transcribe GitHub projects in local audio-to-text fidelity.

Who should use this?

Mac devs in remote pair programming or client calls who use Claude Code or Cursor for AI-assisted coding. Solo creators transcribing YouTube video ideas or live thoughts via microphone to Markdown. Anyone tired of post-meeting TranscribeMe drudgery, especially frontend teams logging auth flow fixes or backend debates.

Verdict

Worth a spin for Mac users needing quick AI transcribe microphone setups—docs are solid, installer painless—but with 82 stars and 1.0% credibility score, it's early-stage; test on non-critical meetings first. Solid local alternative if you're all-in on Sonoma and Whisper.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.