superAIAutoCutVideo  by xiaohu2206

AI video editor automates content creation

Created 7 months ago
285 stars

Top 91.8% on SourcePulse

GitHubView on GitHub
Project Summary

SuperAI影视剪辑 is a lightweight, cross-platform desktop application designed for AI-powered video editing, specifically targeting short dramas and movie commentary content. It offers an out-of-the-box solution that automates video cutting using subtitle and visual analysis, providing a free and efficient workflow for content creators.

How It Works

The application leverages a suite of AI technologies, including visual reasoning (via moondream analysis) for frame analysis, local Automatic Speech Recognition (ASR) for subtitle generation, and Text-to-Speech (TTS) engines like Qwen-tts and edge-tts for narration. Its core workflow supports multi-episode uploads, automatic merging, AI-generated script creation based on subtitles, and subsequent video generation. It also allows for exporting projects to Jianying (CapCut) draft formats.

Quick Start & Requirements

Installation involves separate steps for frontend and backend components. Frontend dependencies are installed via cnpm install after navigating to the frontend directory. The backend requires creating a Python virtual environment (backend.venv or backend/.venv) and installing requirements with pip install -r backend/requirements.txt. The backend is then launched using backend.venv\Scripts\python.exe backend\main.py (Windows) or backend/.venv/bin/python backend/main.py (macOS/Linux).

Prerequisites: Node.js ≥ 18, Python ≥ 3.11, Rust, FFmpeg, and PyTorch (torch). Note: The project explicitly advises against downloading software from GitHub, recommending Baidu Cloud instead. Manual installation of qwen-tts and a specific numpy version (<=2.3.5) is required within the backend virtual environment for Qwen3-TTS functionality.

Highlighted Details

  • Core visual reasoning capabilities powered by moondream analysis.
  • Support for local ASR and multiple TTS engines (Qwen-tts, edge-tts, Tencent TTS).
  • Automated workflow for short drama and movie commentary content creation.
  • Export functionality for Jianying (CapCut) drafts.
  • Multi-project management and customizable prompt/subtitle settings.

Maintenance & Community

The project is under continuous development and optimization, with recent updates focusing on feature enhancements and bug fixes. Users can connect with the developer via Bilibili, Douyin (ID: xiaohu_111), or WeChat (ID: interest_dog) for support or group invitations. The developer notes that the current version is unstable and lacks formal version management, with ongoing intermittent updates.

Licensing & Compatibility

The project is licensed under the MIT license, which generally permits commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

The application is described as unstable and still under intermittent development, with no formal version management. Some features may not be available on all computer configurations. Users unfamiliar with coding might find the setup and usage challenging due to the manual installation steps for certain dependencies like Qwen3-TTS and the command-line-driven nature of the backend. The primary software distribution is recommended via Baidu Cloud, not GitHub releases.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
106 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems") and Jiaming Song Jiaming Song(Chief Scientist at Luma AI).

MoneyPrinterTurbo by harry0703

7.9%
86k
AI tool for one-click short video generation from text prompts
Created 2 years ago
Updated 16 hours ago
Feedback? Help us improve.