AI Subtitle Studio

How to fix broken characters and encoding in subtitle files

When a subtitle file shows é instead of é, ’ instead of an apostrophe, or rows of ??? where Turkish, Chinese or Japanese text should be, the file isn't corrupt — it's being read with the wrong character encoding. AI Subtitle Studio detects the real encoding automatically and rewrites the file as universal UTF-8.

Why subtitle files get garbled

Subtitle files older than the UTF-8 era were saved in regional 8-bit encodings: Windows-1252 for Western Europe, ISO-8859-9 / Windows-1254 for Turkish, GBK/GB18030 or Big5 for Chinese, Shift_JIS for Japanese, EUC-KR for Korean. The file itself doesn't say which encoding it uses — players just guess. Guess wrong, and every non-ASCII character renders as garbage (called mojibake).

Worse, some tools "convert" a misread file and bake the garbage in permanently. That's why fixing encoding should always start from the original file.

Automatic detection and repair

AI Subtitle Studio reads the raw bytes and statistically detects the true source encoding — it recognizes UTF-8 (with or without BOM), UTF-16 LE/BE, Windows-1252/Latin-1, Turkish ISO-8859-9, GB18030/GBK, Big5, Shift_JIS and EUC-KR, among others. It then decodes correctly and saves as UTF-8, the encoding every modern player, TV and streaming server expects. This detection and decoding runs instantly and fully offline — no AI model involved.

If the damage was already baked in by another tool as mojibake, the Cleanup tool's encoding-repair option can additionally reverse the most common broken sequences (like é → é) inside the text itself — also a local, offline pass.

Fix a garbled subtitle file

  1. Open the broken file Drag the .srt/.ass/.sub into AI Subtitle Studio. Detection runs automatically on load.
  2. Verify the preview The editor shows the decoded text — accents, Turkish characters, CJK text should now read correctly.
  3. Save as UTF-8 Export the file. It is written as UTF-8 and will display correctly in VLC, Plex, TVs and browsers.

Repair your subtitle encoding automatically

Free to try on Windows & macOS · full version $9.99 one-time · 100% offline

Download free See pricing

Frequently asked questions

Why do my Turkish subtitles show ý and þ instead of ı and ş?
The file is encoded in ISO-8859-9/Windows-1254 but read as Windows-1252. The app detects the Turkish encoding and converts it to UTF-8, restoring the correct characters.
What encoding should subtitles use today?
UTF-8. It represents every language, and it is the default expectation of modern players and media servers. All files saved by AI Subtitle Studio are UTF-8.
Can encoding issues break subtitle timing too?
No — timing lines are plain ASCII and survive any 8-bit misread. Only the visible text gets garbled, and that is exactly what re-decoding fixes.
The file shows ’ even after conversion. Why?
That indicates double-encoded damage baked in by an earlier tool. Run the Cleanup tool's encoding-repair option, which recognizes and reverses common double-encoding artifacts — entirely offline.

Related guides