Skip to content

Latest commit

 

History

History
122 lines (91 loc) · 3.97 KB

README_EN.md

File metadata and controls

122 lines (91 loc) · 3.97 KB

简体中文 | English

SubErase-Translate-Embed

🎬 All-in-One Video Subtitle Processing Solution

license stars forks

✨ Overview

SubErase-Translate-Embed is a powerful open-source tool that simplifies multilingual video content creation. Using advanced AI technology, it automates the entire process of subtitle detection, removal, translation, and embedding.

Why Choose SubErase-Translate-Embed?

  • 🚀 Fully Automated: One-click solution from subtitle extraction to embedding
  • 🎯 High Accuracy: Precise subtitle recognition powered by PaddleOCR
  • ✨ Perfect Removal: Natural subtitle removal using STTN technology
  • 🌍 Multilingual: Support for multiple language translations
  • 🛠️ User-Friendly: Easy-to-use command-line interface

🎥 Demo

Demo

Left: Original Video | Right: Processed Video

🚀 Key Features

  • Intelligent Subtitle Recognition:

    • High-precision text recognition with PaddleOCR
    • Support for various fonts and complex backgrounds
    • Automatic subtitle position and timestamp detection

  • Professional Subtitle Removal:

    • Smart restoration using STTN technology
    • Seamless removal with natural results
    • Complex background handling

  • High-Quality Translation:

    • Integration with ChatGPT API for accurate translations
    • Preservation of original context and expression
    • Support for multiple language pairs

  • Precise Subtitle Embedding:

    • Customizable subtitle styles
    • Intelligent position adjustment
    • Smooth transition effects

🔧 Installation

Follow these steps to set up SubErase-Translate-Embed:

  1. Clone the Repository:

    git clone --recursive https://github.com/chenwr727/SubErase-Translate-Embed.git

  2. Install Dependencies:

    conda create -n ste python=3.10
conda activate ste
pip install paddlepaddle-gpu==2.6.1.post120 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html
pip install -r requirements.txt

  3. Download Models:

    Save the model files in the ./models directory with the following structure:

    models
├── ch_PP-OCRv4_det_server_infer
└── ch_PP-OCRv4_rec_server_infer
└── sttn.pth

  4. Configuration:

    cp config-template.yaml config.yaml

  5. Install Applications:

    sudo apt install imagemagick
conda install -c conda-forge ffmpeg
conda install -c conda-forge gcc=12.2.0

📖 Usage

Process your video with a single command:

python main.py --video input_video.mp4 --language English

For advanced configuration options, refer to config.yaml.

🤝 Contributing

We welcome all forms of contributions, whether it's new features, bug fixes, or documentation improvements. Please check our contribution guidelines for details.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Thanks to these amazing projects: