Repolode
Submit
zhouxiaoka

zhouxiaoka/autoclip

Open-source desktop tool that turns a link to a long podcast, interview or lecture into short vertical clips with subtitles, covers and publishing copy.

Open-source alternative to:Opus ClipDescript

View on GitHubWebsite
Stars
9,603
Forks
1,721
Open issues
30
Last commit
5 hours ago

What is it?

AutoClip takes a link to a long video, such as a podcast, interview or course, and produces short clips for platforms like Douyin, Xiaohongshu, TikTok, Reels, YouTube Shorts, Bilibili and YouTube. You pick a target platform and the tool scores candidate segments, generating up to 10 qualifying clips per run, each with a cover, title, description, hashtags and a ZIP publishing package.

Editing and rendering run locally on your computer. Subtitles can come from the source video's existing captions, a local Whisper or SenseVoice model, or a configured cloud transcription service. Analysis can use cloud LLM providers via API key, or local models through Ollama and LM Studio. Speaker framing follows the person talking, and an optional editor is available for adjustments.

Beyond the desktop app, the project offers a Docker-based web deployment and a CLI with an MCP server, so the same clipping pipeline can be driven from scripts or AI agents.

Key features

  • 01Generates short clips from long videos for multiple vertical and horizontal platform formats.
  • 02Creates a cover, title, description and hashtags for each clip, exported as a ZIP package.
  • 03Transcribes with existing subtitles, local Whisper or SenseVoice, or configured cloud services.
  • 04Supports cloud LLM APIs and local models via Ollama and LM Studio.
  • 05Keeps editing and rendering on the local machine.
  • 06Provides a desktop app, Docker web deployment, and CLI/MCP interface.

Use cases

  • Repurposing podcast or interview episodes into short clips for Douyin, Xiaohongshu or TikTok.
  • Turning long course or lecture recordings into short promotional videos for YouTube Shorts or Reels.
  • Batch-processing video content from an AI agent or script through the CLI or MCP server.

Topics

#video-editing#video-clipping#subtitles#ai-video#content-repurposing#ffmpeg#whisper#desktop-app#ai-agent#ai-video-editor

FAQ

What is autoclip?
Open-source desktop tool that turns a link to a long podcast, interview or lecture into short vertical clips with subtitles, covers and publishing copy. AutoClip takes a link to a long video, such as a podcast, interview or course, and produces short clips for platforms like Douyin, Xiaohongshu, TikTok, Reels, YouTube Shorts, Bilibili and YouTube. You pick a target platform and the tool scores candidate segments, generating up to 10 qualifying clips per run, each with a cover, title, description, hashtags and a ZIP publishing package.
Is autoclip free to use? What license does it have?
autoclip is open source under the MIT license. Check the license terms before using it in commercial products.
Is autoclip still maintained?
Actively maintained. The last commit to autoclip was on Oct 21, 2026.
What can I use autoclip for?
Repurposing podcast or interview episodes into short clips for Douyin, Xiaohongshu or TikTok. Turning long course or lecture recordings into short promotional videos for YouTube Shorts or Reels. Batch-processing video content from an AI agent or script through the CLI or MCP server.
Where can I get autoclip?
The source code and installation instructions are on GitHub: https://github.com/zhouxiaoka/autoclip

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects