Drop in a bilingual Japanese + English SRT and this tool splits it into a Japanese-only file and an English-only file — download either as SRT or as plain TXT. It runs entirely in your browser, so your subtitle file is never uploaded. You don't pick a language: it detects Japanese lines by their characters (both kana and kanji) and separates them from the English lines, even when the line order is inconsistent.
— or paste the subtitle text below —
The tool parses your SRT into individual cues, then reads each text line and decides whether it is Japanese or English by the characters it contains. Lines with Japanese characters — hiragana, katakana, or kanji — go into the Japanese track; lines in the Latin alphabet go into the English track. The timestamps are copied unchanged, so both output files stay perfectly in sync with the video.
Because it classifies by script rather than by line number, it still works when the English line comes first, when a cue has only one language, or when a file mixes patterns. A Japanese line written entirely in kana is detected correctly — not just lines that contain kanji. If auto-detection ever gets it wrong, switch to By line position mode.
Yes. Each side has its own download buttons. Export as SRT to keep the timestamps for a video editor or player, or export as TXT to get a clean, timecode-free transcript in a single language. The TXT option is handy for study material, translation checks, or when you only need the Japanese script or only the English script as plain text.
No. This tool runs 100% in your browser — your subtitle file never leaves your device and is never sent to a server. It works offline once the page has loaded, and there is no upload size limit. This is the same local-first approach behind GeekLink, our desktop subtitle app.
Yes. The tool recognizes hiragana and katakana as well as kanji, so a Japanese line with no kanji is still routed to the Japanese track. Only lines written in the Latin alphabet go to the English side.
No. It detects each line automatically by its characters, so you only provide the file. If your file has an unusual layout, switch to "By line position" mode, which sends the first line of every cue to the Japanese file and the rest to the English file.
Yes. Timestamps are copied unchanged into both output files, so the Japanese-only and English-only subtitles stay perfectly in sync with the original video.
This splitter works on subtitle files you already have. If the subtitles are burned into the picture (hardcoded) with no SRT, you need OCR extraction first. GeekLink, our desktop app for Mac and Windows, extracts hardcoded subtitles via OCR, transcribes speech, translates, and burns subtitles in — including batch processing across many videos.
Splitting a different language pair? Try the Chinese-English splitter or the Korean-English splitter.
This free tool splits files you already have. GeekLink extracts hardcoded subtitles via OCR, transcribes and translates with AI, and burns subtitles into video — batch-processing dozens of videos, all locally on your Mac or Windows PC.
Try GeekLink Free