Encoding Fixer (Multilingual Mojibake Repair)Auto-detect and fix garbled text caused by wrong encoding

What actually causes gibberish like 'é' or a wall of random symbols?

Mojibake (garbled text) almost always comes from decoding text with the wrong character encoding. A Traditional Chinese file saved in Big5, if opened as UTF-8 by mistake, has every Chinese character split into several meaningless symbol clusters. Conversely, a UTF-8 file opened as Windows-1252 or Latin-1 produces things like 'é' where you'd expect 'é'. This tool works backwards from that: it takes your garbled text or file and tries re-decoding it through 20+ common encodings, checking which result reads like real, natural-language text, then ranks the candidates by confidence so you can pick the right one.

How does the tool decide which candidate is correct?

For each encoding it tries, the tool analyzes the resulting text: how many unrecognizable replacement characters (like �) show up, whether the character distribution matches a plausible language, whether punctuation looks sane, and so on — combining these signals into a confidence score, with the highest-scoring candidate shown first. Supports encodings for all 18 languages: Traditional Chinese (Big5, CP950), Simplified Chinese (GBK, GB2312, GB18030), Japanese (Shift-JIS, EUC-JP), Korean (EUC-KR, CP949), Cyrillic (Windows-1251, KOI8-R), Arabic (Windows-1256), Hebrew (Windows-1255), and ISO-8859 / Windows-125x series for European languages.

Upload a file, or paste text — which should you use?

If you have the original file (.txt, .csv, .log, etc.), use 'Upload File' mode — the tool can read the raw bytes directly, giving the most accurate recovery. If you're just pasting text that's already garbled on your screen (copied from somewhere else), some encoding information may already be lost in that copy — in that case use 'Paste Text' mode, which works well for Latin-based UTF-8 mojibake ('é'-style errors) but has limited ability to recover CJK (Chinese/Japanese/Korean) text that's already been copy-pasted. The tool shows up to 5 ranked fix candidates; choose the correct one and copy or download.

Once text has been copied and pasted (rather than read directly from the original file's bytes), some encoding information may be permanently lost — even trying every encoding won't always achieve a perfect recovery. Upload the original file whenever possible for the best result.