Subtitle Encoding Fixer: Repair Garbled Subtitles to Clean UTF-8

Free browser tool · Nothing uploaded · Updated August 3, 2026

Drop in a subtitle file that shows up as garbled text (锟斤拷, 卯碌忙, ä½ å¥½, or □□□) and this tool detects its real encoding and re-saves it as clean UTF-8. The text was never broken — it was just stored in an older encoding (like GB18030 or Big5) and read with the wrong one. Everything runs in your browser, so your file is never uploaded.

🔒 100% local. Your file is processed in your browser and never uploaded.
Drop your subtitle file here, or click to choose
.srt, .vtt, .ass, .ssa, .sub — the file is read as raw bytes on your device.

Why do subtitles show up as garbled text?

A subtitle file is just text stored as bytes, and those bytes have to be read with the same "alphabet" (encoding) they were written in. Most garbled Chinese subtitles are stored in GB18030 or Big5 but get read as UTF-8, so every character comes out wrong. The words are fine — only the reading key is mismatched. This tool reads the file as raw bytes, figures out which encoding it actually is, and rewrites it in UTF-8, the format every modern player and editor understands.

How does the encoding detection work?

The tool first checks whether the file is already valid UTF-8. If it is not, it decodes the bytes with each likely encoding — GB18030, Big5, Shift_JIS, EUC-KR, Windows-1252 — and scores the results, keeping the one that produces the most real characters and the fewest garbage symbols. It also repairs "double-encoded" text (mojibake), the kind where 你好 turns into ä½ å¥½. If the automatic guess is wrong, you can pick the encoding yourself and the preview updates instantly.

Is my subtitle file uploaded anywhere?

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, with no file-size limit. This is the same local-first approach behind GeekLink, our desktop subtitle app.

Frequently Asked Questions

Why does my Chinese subtitle look like 锟斤拷 or a row of boxes?

The file is almost certainly stored in a Chinese encoding such as GB18030 or Big5, but your player is reading it as UTF-8. The text is intact; the reading key is wrong. This tool detects the real encoding and re-saves the file as UTF-8 so it displays correctly.

Can it fix text like "ä½ å¥½" (mojibake)?

Yes. That pattern is double-encoded text, where UTF-8 bytes were read as Latin-1 and saved again. The tool detects this case and reverses it to recover the original characters.

Do I have to know the original encoding?

No. Detection is automatic. If the guess is wrong, a dropdown lets you try GB18030, Big5, Shift_JIS, EUC-KR, or Windows-1252, and the preview updates so you can confirm before downloading.

Why can't I just paste the text instead of uploading a file?

Encoding problems live in the file's bytes. Once text is pasted, it has already been decoded (often incorrectly and unrecoverably), so the fix has to work from the original file. That is why this tool takes a file rather than pasted text.

Need to split a bilingual subtitle instead? Try the Chinese-English splitter, or see all free tools.

Working with subtitles at scale?

This free tool fixes 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