# YaoFANGUK/video-subtitle-extractor

视频硬字幕提取，生成srt文件。无需申请第三方API，本地实现文本识别。基于深度学习的视频字幕提取框架，包含字幕区域检测、字幕内容提取。A GUI tool for extracting hard-coded subtitle (hardsub) from videos and generating srt files.

Repository: https://github.com/YaoFANGUK/video-subtitle-extractor
Canonical: https://ross.abutalabs.com/products/video-subtitle-extractor
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, ocr, subtitles, srt, hardsub, extract, ripper, subrip
Last push: 2026-04-09T15:27:37+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 76, release rhythm 46, longevity 100
- inputs: {"age_days": 2159, "days_push": 146, "days_rel": 151, "gap_med": 345, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 9401, forks 948 (observed 2026-08-28T04:10:31.023600+00:00)

## What it is
A GUI application that extracts hard-coded (burned-in) subtitles from videos and generates SRT subtitle files using local deep-learning-based OCR. It detects subtitle regions in video frames, recognizes text in 87 languages, and deduplicates lines without requiring any third-party OCR APIs.

## Use cases
- extract hard-coded subtitles from a video to an srt file
- rip burned-in subtitles without uploading video to online OCR services
- batch extract subtitles from multiple videos
- generate srt or txt files from hardsub videos
- recognize Chinese, English, Japanese, Korean and other subtitles locally
- remove watermarks and channel logos text from extracted subtitles

## When to choose
- you need subtitles from videos with burned-in text and no external subtitle file exists
- you want fully local OCR with no third-party API keys or cloud uploads
- you need multilingual subtitle extraction including CJK languages
- you want GPU-accelerated batch processing with a GUI

## When to avoid
- the video already has soft/embedded subtitle tracks you can extract directly with ffmpeg
- you need real-time subtitle extraction during playback
- you cannot install Python/CUDA or run heavy deep-learning models on your machine
- you need frame-accurate results on CPU only - precise mode is very slow

## Facets
- artifact type: application
- maturity: active
- function: ocr, video-processing, machine-learning, gui
- domain: media, deep-learning
- platform: windows, python
- tags: hardsub, subtitles, srt, subtitle-extraction, video-to-text, offline-ocr, video, natural-language-processing, macos, linux, gpu

## Member repositories
- YaoFANGUK/video-subtitle-extractor (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:31.023600+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:22:11.101747+00:00, confidence not recorded.
  - readme: https://github.com/YaoFANGUK/video-subtitle-extractor (fetched 2026-08-28T04:10:31.023600+00:00, sha 4008be862cfb)
- Data as of 2026-08-30T08:39:29.467469+00:00.