Fix Garbled Subtitles and Convert to UTF-8
Question marks, boxes, “é” or random Chinese characters instead of accents? Your subtitle is in a different encoding from the one your player expects. Fix it in seconds.
Try other encodings — click the one that looks right
🔒 Runs in your browser. Your files and text are not uploaded.
How to use the Subtitle Encoding Fixer
- Drop the garbled subtitle file. It is read as raw bytes, so no information is lost.
- Leave Source encoding on Auto-detect. The preview shows the first lines; if they look wrong, click another encoding in the “Try other encodings” list until the text reads correctly.
- If the text is already garbled in the file itself (you can see “é” or “浣犲ソ” in a text editor), change What is wrong? to “The text is already garbled” and choose how it was misread.
- Download the UTF-8 file. Add the BOM only if an old player refuses plain UTF-8.
Why subtitles break
A text file is just bytes. The encoding is the table that says which byte means which letter. Subtitles from older sources use national code pages — Windows-1252 for Western Europe, Windows-1250 for Central Europe, Windows-1251 for Cyrillic, Windows-1256 for Arabic, GBK for Simplified Chinese, Big5 for Traditional Chinese, Shift-JIS for Japanese, EUC-KR for Korean. Modern players assume UTF-8, so bytes are shown as the wrong letters. The reverse also happens: a UTF-8 file is opened as Windows-1252 and every é becomes é.
Recognising the symptoms
| You see | What happened | Fix |
|---|---|---|
é, ä, ’ | UTF-8 text read as Windows-1252 | “Already garbled” → misread as Windows-1252 |
浣犲ソ, 鎴戜滑 | UTF-8 Chinese read as GBK | “Already garbled” → misread as GBK |
привет | UTF-8 Cyrillic read as Windows-1252 | “Already garbled” → Windows-1252 |
é (even more strange) | Garbled twice | Run repair twice |
? or � replacing letters | Characters were lost when the file was saved | Cannot be repaired — find another copy |
Which encodings are detected
The detector examines the byte patterns and the letters that appear, and ranks the candidates: UTF-8, UTF-16, GBK / GB18030, Big5, Shift-JIS, EUC-JP, EUC-KR, Windows-1250 to 1258, Windows-874 (Thai), KOI8-R, DOS-866 and ISO-8859-2/15. Short files with few accented letters can be ambiguous, which is why the preview and the clickable list let you decide the last step yourself.
Tips for avoiding the problem
- Always save subtitles as UTF-8 in your editor (“Save as” → encoding).
- Do not open and re-save a subtitle in a program that guesses the encoding wrongly.
- For Windows Notepad, choose “UTF-8” (not “ANSI”) in the Save dialog.
- Keep a backup of the original file before you repair it.
Frequently asked questions
What is the difference between “different encoding” and “already garbled”?
In the first case the file is in a legacy encoding such as Windows-1251 and shows wrong characters in your player. In the second case someone already converted the file incorrectly, so the garbled letters are now stored in the file itself (often as UTF-8) and have to be reversed.
Can question marks be repaired?
No. If the file contains literal “?” characters instead of letters, the information is gone. Look for another copy of the subtitle.
Which encoding should I choose for Chinese subtitles?
Auto-detect usually decides between GBK/GB18030 (Simplified), Big5 (Traditional) and UTF-8 correctly. If it gets stuck, try each in the list.
Should I add a BOM?
Only for old Windows programs that otherwise treat UTF-8 as ANSI. Most players handle UTF-8 without it.
Will timing and tags be changed?
No. Only the text encoding changes. You can also convert line endings.
Does the file get uploaded for detection?
No. Detection runs in JavaScript in your browser.