Project Title

video-subtitle-extractor — A Deep Learning-Based GUI Tool for Extracting Hard-Subtitles from Videos and Generating SRT Files

Overview

Video-subtitle-extractor (VSE) is a Python-based open-source tool designed to extract hard-coded subtitles from videos and generate SRT files locally without the need for third-party APIs. It stands out for its deep learning framework that includes subtitle area detection and content extraction, supporting a wide range of languages and offering multiple extraction modes for different accuracy and speed requirements.

Key Features

Local OCR text recognition without the need for external APIs
Supports subtitle extraction in 87 languages, including Chinese, English, Japanese, Korean, and more
Offers three extraction modes: Fast, Auto, and Accurate, catering to different user needs
Batch subtitle extraction capability for processing multiple videos
GUI tool for easy adjustment of subtitle areas and extraction process management

Use Cases

Video editors and content creators needing to extract subtitles for editing or translation purposes
Subtitlers who want to convert hard subtitles into editable SRT files
Educational institutions and language learners using videos for language study, needing subtitles for better comprehension

Advantages

High accuracy and speed in subtitle extraction, especially when utilizing GPU acceleration
Customizable text filtering and replacement for removing watermarks or specific texts
Supports a wide range of video resolutions and subtitle areas, accommodating various video formats

Limitations / Considerations

The project is not licensed under a known open-source license, which might affect its usage in commercial projects
The "Accurate" mode is significantly slower than the other modes, which could be a limitation for large-scale projects with tight deadlines
The tool may not work well with video and program paths containing Chinese characters or spaces, which could require additional setup or adjustments

Subtitle Edit: A cross-platform open-source subtitle editor that supports a wide range of formats, but does not focus on hard-subtitle extraction.
Aegisub: A free, open-source subtitle editor that provides powerful tools for styling and time coding, but does not specialize in hard-subtitle extraction from videos.
ffmpeg: A comprehensive multimedia framework that can handle video and audio processing, including subtitle extraction, but lacks the GUI and deep learning capabilities of VSE.

Basic Information

GitHub: https://github.com/YaoFANGUK/video-subtitle-extractor
Stars: 8,045
License: Unknown
Last Commit: 2025-11-13

📊 Project Information

Project Name: video-subtitle-extractor
GitHub URL: https://github.com/YaoFANGUK/video-subtitle-extractor
Programming Language: Python
⭐ Stars: 8,045
🍴 Forks: 835
📅 Created: 2020-10-04
🔄 Last Updated: 2025-11-13

🏷️ Project Topics

Topics: [, ", d, e, e, p, -, l, e, a, r, n, i, n, g, ", ,, , ", e, x, t, r, a, c, t, ", ,, , ", h, a, r, d, s, u, b, ", ,, , ", o, c, r, ", ,, , ", r, i, p, p, e, r, ", ,, , ", s, r, t, ", ,, , ", s, u, b, r, i, p, ", ,, , ", s, u, b, t, i, t, l, e, s, ", ]

🎥 Video Tutorials

去除水印、台标文本、原视频硬字幕，可配合：video-subtitle-remover (VSR)

This article is automatically generated by AI based on GitHub project information and README content analysis

video-subtitle-extractor

Project Description