YuzuLingo vs RapidVideOCR (SWHL)
YuzuLingo stays in the browser. You cover burned-in subtitles on the video you are already watching, including YouTube, Bilibili, and other web players. There is no script to run and no video file to start from. Turn on Interact to read the burned-in line on your device, with furigana or pinyin when the language has them. Show that text always, on hover, or while you hold a hotkey.
RapidVideOCR is SWHLâs Python package, in the repo RapidVideOCR. You install it with pip. The rapid_videocr command reads a folder of RGBImages or TXTImages that VideoSubFinder already saved, and RapidOCR writes an SRT, an ASS, and a TXT. It does not open the video itself. A script in the repo can call VideoSubFinderWXW.exe first if you give it that exe path. The desktop app is a separate repo, RapidVideOCRDesktop. There is a hosted demo on Hugging Face. RapidOCR runs on the device, and its language list includes English, Spanish, Japanese, Korean, and Chinese in simplified and traditional, plus other languages on that list. Release v3.1.1. The last commit was about a year ago. Apache-2.0.
GitHub - SWHL/RapidVideOCR: đŠ Extract video hard subtitles and automatically generate corresponding srt files.Where each one fits
| Feature | YuzuLingo | RapidVideOCR |
|---|---|---|
| Install in the browser, no script to run | ||
| Cover burned-in subs while a web video plays | ||
| No video file to start from | ||
| Read that burned-in line while the web video plays, on your device | ||
| Furigana or pinyin on that line | ||
| Show that text always, on hover, or while you hold a hotkey | ||
| Turn VideoSubFinder frames into an SRT, ASS, or TXT | ||
| Read English, Spanish, Japanese, Korean, and Chinese | ||
| OCR stays on the device | ||
| Add the line to an Anki deck |