OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Viralcutter

ViralCutter is an open-source Python application that processes long YouTube videos into short vertical clips, combining AI-assisted moment selection with transcription, dynamic captions, face tracking and subtitle translation. It runs locally and offers a Gradio WebUI and a command-line entry point.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Viralcutter — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to turn long YouTube footage into vertical short-form edits on your own computer, with options for online AI services or local GGUF models.

Who it is for

  • Video creators repurposing long YouTube content for TikTok, Instagram Reels or YouTube Shorts.
  • Users who want local video processing and are comfortable installing software and configuring AI services or local models.

Documented uses

  • Find candidate highlights in a long video and produce short clips.
  • Transcribe speech and create customizable, dynamically highlighted captions.
  • Reframe horizontal footage to 9:16 and use face-aware layouts, including split screen for two faces.
  • Generate translated subtitles for localized short clips.
Before you choose

Requirements & limits

The project is presented as open source under GPL-3.0 and provides local installation instructions. Its standard setup references online AI services; an advanced installer supports local GGUF models.

  • The project is a local application, so setup and processing depend on the user’s computer.
  • The README recommends Python 3.10 or 3.11, FFmpeg and NVIDIA drivers for CUDA acceleration; an NVIDIA GPU is strongly recommended for speed and local AI.
  • Gemini configuration uses an API key in api_config.json; local GGUF models are placed in the models/ directory.
  • Active-speaker camera switching is labeled experimental, and Premiere Pro XML export is labeled beta.
  • WhisperX is identified as the transcription basis; the README also describes dynamic word-level caption styling and automatic 9:16 cropping.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and launch ViralCutter on Windows

Set up the project and open its WebUI using the repository’s documented Windows instructions.

Before you start

  • Python 3.10.x or 3.11.x, with Add Python to PATH enabled
  • FFmpeg
  • Visual Studio C++ Build Tools with Desktop development with C++, Windows 10/11 SDK and MSVC v143; listed for compiling InsightFace
  • Updated NVIDIA drivers are recommended for CUDA acceleration; an NVIDIA GPU is strongly recommended for speed and local AI
  • The ViralCutter repository and its installation scripts

Steps

  1. Install the listed prerequisites, including FFmpeg and the applicable C++ build components.

  2. From the ViralCutter folder, run install_dependencies.bat for the standard setup, or install_dependencies_advanced_LocalLLM.bat for the local-LLM setup.

  3. For the standard option, add a Gemini key to api_config.json if using Gemini. For local GGUF models, place model files in the models/ directory.

  4. Launch the interface with run_webui.bat, or use python main_improved.py for the CLI.

Expected outcome

The repository’s instructions lead to the WebUI or CLI entry point after dependency installation and any chosen AI configuration.

Guide limitations

  • The advanced local-LLM installation is described as requiring a capable GPU and C++ Build Tools.
  • The README recommends NVIDIA hardware for speed and local AI; performance depends on the user’s setup.
  • These are published instructions, not a verified installation or compatibility guarantee.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 October 2026
Original website title
RafaelGodoyEbert/ViralCutter · Official homepage title · retained in its source language
How we review sourcesSuggest a correction