Skip to content

FunClip

Clip video by picking recognised text or a speaker

Visit official site

What FunClip is

The tool is described as a fully open-source, locally deployed automated video editing tool that runs speech recognition with the FunASR Paraformer series. It integrates Paraformer-Large and CAM++ speaker recognition, predicts timestamps in one pass, and lets users pick text fragments or speakers to clip the matching video.

What you can do with it

  • Clip video by selecting recognised text fragments
  • Clip the passages of a chosen speaker
  • Cut multiple segments freely and generate SRT subtitles automatically
  • Add hot words to improve recognition of entity and personal names

Application scenarios

  • Education providers clip key passages from teaching videos
  • Video bloggers make content containing specific dialogue
  • Companies cut chosen parts of product introduction videos for marketing

Getting started

  • Clone the FunClip repository to a local machine
  • Install the required Python dependencies
  • Install ImageMagick when automatic subtitle generation is needed
  • Start the Gradio service locally and use the interface in a browser

Pros & cons

✓ What we like

  • The tool is fully open source and runs locally to protect privacy
  • Gradio interaction is described as simple and easy to use

! What to watch out for

  • Local use starts with cloning the repository and installing Python dependencies
  • Automatic subtitle generation needs ImageMagick installed

FAQ

How is a clip produced?

Users select text fragments or speakers from the recognition results and click the clip button.

Does it run locally?

The documented steps clone the repository and start the Gradio service on a local machine.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review