denven

denven / mediascribe

Public

CLI for audio, video, and text transcription with ASR providers and LLM-powered summarization via local or cloud backends.

asr audio-transcription cli ffmpeg litellm
10
0
100% credibility
Found Apr 13, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

MediaScribe is a command-line tool for transcribing audio, video, and text files into structured summaries using local or cloud-based speech recognition and language models.

How It Works

1
🔍 Discover MediaScribe

You hear about this friendly tool that turns your meeting recordings, videos, or notes into clear, structured summaries just like magic.

2
💻 Set it up on your computer

You grab the tool and get it ready with simple steps, connecting any smart helpers if you want faster results.

3
📂 Pick your file or folder

You choose an audio clip, video from your drive or online link, or a bunch of notes to process.

4
✨ Start the summary magic

You tell the tool to go, and it smartly pulls out words from sound or screen text while you relax.

5
📝 It handles subtitles first

For videos, it grabs ready-made captions if available, or listens closely to the audio as backup.

✅ Enjoy your perfect summary

You get a neat document with key points, topics, action items, and decisions, ready to share or act on.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 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 mediascribe?

MediaScribe is a Python CLI that transcribes audio clips, videos, and text files using local or cloud ASR backends like Whisper, Azure, Aliyun, or iFlytek, then generates LLM-powered summaries via Ollama or cloud providers. It solves the hassle of piecing together separate tools for media processing by offering end-to-end workflows: feed in a meeting recording or YouTube URL, get timed transcripts with speaker labels and structured Markdown summaries. Commands like `mediascribe meeting.wav --asr azure` or `mediascribe video https://youtube.com/watch` handle local files, directories, or remote links with yt-dlp support.

Why is it gaining traction?

It stands out with subtitle-first video processing that falls back to audio extraction only when needed, plus staged reusability—transcribe first with `mediascribe-transcriber`, summarize later via `mediascribe-text`. Local-first options keep costs down without sacrificing diarization or quality, and provider abstraction lets you swap ASR/LLMs via env vars or flags. Devs dig the clean outputs preserving metadata for pipelines.

Who should use this?

DevOps engineers batch-processing meeting audio on Linux servers, content creators summarizing Bilibili/YouTube videos before editing, or AI devs prototyping mediascribe ai workflows with CLI github actions. Ideal for teams handling multilingual podcasts or lectures needing quick key points and action items.

Verdict

Try it for lightweight media transcription pipelines—docs are solid with benchmarks and hardware guides, but at 10 stars and 1.0% credibility, it's early-stage; expect tweaks for production scale. Pair with uv for fast installs.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.