autogame-17

Scribe — multi-platform video → local Whisper transcription → LLM proofread with Typeless-style glossary learning. Wails v2 desktop app for macOS & Windows.

14
4
69% credibility
Found May 18, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Scribe Studio is a desktop application that transforms videos from various platforms into clean, searchable text. You paste a video link, it downloads and transcribes the audio using local AI, has an AI assistant proofread and improve the text, builds your personal terminology list over time, and lets you export the final transcript. It's like having a smart assistant who watches videos and takes notes for you.

How It Works

1
🎬 You have a video worth keeping

You watched something great on video platforms and want to save it as text you can search, edit, and share.

2
🔗 You paste any video link

Drop in a link from YouTube, Bilibili, WeChat Channels, or dozens of other platforms. Scribe handles the rest.

3
🎧 Your video gets converted to audio

Scribe extracts just the audio, then uses a local AI model (Whisper) to convert speech to text automatically.

4
AI polishes your transcript
📚
You build a personal glossary

Accept good suggestions and your special terms get remembered for all future transcripts.

💾
You export your finished text

Download as markdown for documents, SRT for captions, or plain text for any purpose.

You have searchable, editable text

Your video content is now readable text you can search, quote, or embed anywhere. Job done.

Sign up to see the full architecture

3 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 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 scribe-studio?

Scribe-studio is a desktop application that turns video content into searchable, editable text transcripts. Built with Go and Wails v2, it downloads videos from WeChat Channels, YouTube, Bilibili, X, Douyin, TikTok, and over 1700 other platforms via yt-dlp, then runs local Whisper transcription with optional LLM proofreading using Claude or Gemini. The standout feature is its glossary learning system that adapts to your terminology over time, improving accuracy with each use. It exports to markdown and SRT formats.

Why is it gaining traction?

The Typeless-style incremental learning is the hook -- instead of manually managing terminology lists, the app learns from your corrections and applies them automatically. The WeChat Channels integration with built-in MITM proxy is unique for Chinese content creators. Local Whisper transcription means no API costs or privacy concerns, while the LLM proofreading layer adds quality without requiring manual editing. The unified pipeline from download through transcript with glossary feedback creates a smooth workflow that alternatives don't offer.

Who should use this?

Researchers and journalists working with Chinese video content will benefit most from the WeChat Channels integration. Content creators who need searchable transcripts of multi-platform videos. Developers building workflows around video-to-text pipelines. Anyone frustrated with per-session API costs for transcription will appreciate the local-first approach.

Verdict

With only 14 stars and a credibility score of 0.699%, this is an early-stage project that needs community validation before production use. The feature set is solid and the architecture makes sense, but limited documentation and small user base mean you're partly beta-testing. Worth watching, but wait for a 1.0 release or significant community growth before betting a workflow on it.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.