Why subtitles show strange characters, and how to fix the encoding

What an encoding is

A subtitle file is text, and text is stored as numbers. The encoding is the table that says which number means which letter. Today almost everything uses UTF-8, which covers every writing system. Older subtitle files were often saved with a regional table instead, such as Windows-1252 for Western European languages, Windows-1251 for Cyrillic or Windows-1256 for Arabic.

When a player reads a file with the wrong table, the numbers turn into the wrong letters. The words are still in the file; they are only being read the wrong way.

Recognizing the problem

If é shows as é or ü shows as ü, a UTF-8 file is being read as Windows-1252. If letters show as � or as boxes, a file saved in an older regional encoding is being read as UTF-8. If a whole line turns into unrelated accented letters, the file was probably made for another language group, for example Cyrillic subtitles read with a Western table.

Fixing it

When you open a file on this site, the page checks for UTF-8 and UTF-16, recognizes Cyrillic text saved as Windows-1251, and otherwise assumes Windows-1252. If the preview still looks wrong, choose the language group the subtitles were made for in the Text encoding list, such as Cyrillic, Greek, Turkish, Arabic or Japanese, and the preview updates immediately.

Every file the tools create is saved as UTF-8, so once the text looks right in the preview, the downloaded file will show correctly in modern players, editors and browsers.

If a player still shows the wrong letters

Some older players and TVs ignore UTF-8 unless they are told about it. Look for a subtitle encoding or character set setting and choose UTF-8. If the problem is a missing font rather than an encoding, the letters will show as empty boxes even in the correct encoding; choosing another subtitle font in the player usually helps.