Clipping
ViralMint
Local-first video pipeline: trend scouting, clipping, captioned shorts.
★ 122 starsAGPL-3.0
View guide →FunASR-powered video transcription, subtitles and LLM-assisted clipping with a local Gradio UI.
FunClip is an open-source clipping project by modelscope (modelscope/FunClip on GitHub). It has 6,379 GitHub stars and 756 forks, is released under the MIT license, and was last updated on GitHub on 8 October 2026.
Gather the videos you want to transcribe or cut, and decide whether you need subtitles, speaker-specific clips or highlight reels.
Clone the repo, create the documented Python 3.12 virtual environment and install the requirements.
Pick a model — the default Paraformer pipeline, Fun-ASR-Nano, SenseVoice or the MOSS backend — and let the weights download.
Launch the Gradio UI, upload a video, copy the transcript segments you want, and export the clip with subtitles.
Integrates Alibaba's Paraformer-Large, one of the best-performing open-source Chinese ASR models, plus Fun-ASR-Nano and SenseVoice options.
Uses SeACo-Paraformer hotwords so names and domain terms you specify are recognized more accurately.
The CAM++ speaker recognition model lets you clip the segments spoken by a specific speaker.
Select the transcript text you want and FunClip cuts those segments, returning full-video SRT and target-segment SRT subtitles.
Everything runs through a browser UI that you can host on your own server; no video leaves your machine unless you choose a cloud LLM gateway.
Optional LLM integration helps pick and refine clip segments from the transcript.
Clone FunClip and create the documented Python environment.
git clone https://github.com/modelscope/FunClip.git
cd FunClip
python3.12 -m venv .venv && . .venv/bin/activateInstall the Python requirements and verify the environment.
python -m pip install -r requirements.txt
python -m pip checkStart FunClip locally and open the interface in your browser.
python funclip/launch.pyFunClip is a modelscope project, built on Alibaba's FunASR and Paraformer model family.
The code is MIT licensed and free; the ASR model weights are downloaded separately and governed by the terms on their model pages.
Core transcription and clipping run locally with downloaded open models — no key needed. Optional LLM gateways for assisted clipping use your own provider key.
FunClip is built around Paraformer/FunASR, which is especially strong for Chinese speech, and adds hotwords and speaker-based clipping in one Gradio tool rather than a bare transcription script.
Clone the repo, create the venv, install requirements, run python funclip/launch.py and upload a video.
Local-first video pipeline: trend scouting, clipping, captioned shorts.
AI-native, local-first video editor by Ribbi — runs entirely in the browser. Rust/WASM engine, WebGPU compositing, WebCodecs export; humans and LLM agents ed…
Desktop highlight editor with Whisper transcription and LLM clip scoring.
Local-first clipper: Whisper plus Ollama turn long video into captioned 9:16 clips.
AI-powered video editor that turns raw footage and a creative brief into a polished ad using an ensemble of AI agents (Google Gemini + FFmpeg)
Gausian - Rust-based local video editor for AI video production
The most-starred other Clipping projects in the LibreAI catalogue — the realistic shortlist if FunClip is not the right fit.
WhisperX — Whisper transcription with word-level timestamps and speaker diarization: 70x realtime batched inference, forced phoneme alignment, BSD-2-Clause
Toonflow 是开源 AI 创作平台，融合无限画布、AI Agent 与可视化工作流，支持图像生成、视频生成、智能分镜及短剧创作。支持本地部署、自由接入模型，提供跨平台桌面端，并可通过 MCP 与插件扩展创作能力。Open-source AI creative…
🚀 Truly open-source AI avatar(digital human) toolkit for offline video generation and digital human cloning
macOS video editor built for AI
MoneyPrinter — automate YouTube Shorts from a single topic: local Ollama writes the script and metadata, stock footage and voiceover are assembled into a finished short, MIT
Free and open — grab the source, star the repo, and start creating.