Fix subtitle encoding and save it as UTF-8.
To fix a subtitle's encoding, drop the SRT file. We work out which encoding it was saved in, such as Windows-1252, ISO-8859-1 or UTF-16, repair text where é stands in for é, and show every line before and after. You download clean UTF-8, with or without a BOM, ready for your TV.
Free, no sign-up- Encoding found
- Line endings
Tired of fixing other people's subtitles? We make the right subtitles straight from your video.
Lines fixed
Before: how a player expecting UTF-8 shows the file. After: the text that goes into the new file.
| Line | Before | After |
|---|
How to fix subtitle encoding
- 01
Drop the file
SRT, VTT, ASS, SBV, SUB or TXT. Use the file rather than copied text, because the encoding lives in the bytes.
- 02
Check before and after
We show the encoding found, the line endings and every line that changes, as your TV showed it and as it will read.
- 03
Download as UTF-8
With or without a BOM, with CRLF or LF line endings. It keeps the same file name, ready to sit next to the video.
Why subtitles show ? or à instead of accented letters
Every text is stored as numbers, and the encoding is the table that says which number is which letter. The letter é, for example, is one byte in Windows-1252 and two bytes in UTF-8. When subtitles were saved with one table and the player reads them with another, letters come out wrong. Spanish, French, Portuguese and German use accents constantly, so the problem shows up in almost every line.
There are two different symptoms. If you see ? or a diamond with a ?, the file is in Windows-1252 or ISO-8859-1 and the player expects UTF-8. If you see é instead of é and ñ instead of ñ, the opposite happened: a UTF-8 file was opened as Windows-1252 and saved that way, sometimes more than once. We fix both.
| You see | Should be | What happened |
|---|---|---|
| Caf� con a�o | Café con año | Windows-1252 file read as UTF-8 |
| Café con año | Café con año | UTF-8 read as Windows-1252 and saved again |
| Café | Café | The same mistake, twice |
| “Hello†| “Hello” | Curly quotes from misread UTF-8 |
| Caf? | Café | Accent lost earlier: cannot be recovered |
How we find the file's real encoding
First we look for the marks at the start of the file that identify UTF-8 with a BOM and UTF-16. With no mark, we test whether the bytes form valid UTF-8; a Windows-1252 file with accents almost never does. If it is not UTF-8, we tell Windows-1252, ISO-8859-1 and ISO-8859-15 apart by the bytes only each one uses, such as Windows curly quotes and the euro sign.
Then we look for text garbled in two layers, like é and “, and rebuild each letter, line by line. Lines that were already right stay as they are, so a file that is half fine and half broken comes out whole. Before you download, you see a list of every line that changes.
UTF-8, UTF-8 with BOM or ANSI: which to pick for TVs, players and editors
UTF-8 is the right choice almost every time: YouTube, VLC, Premiere, DaVinci Resolve, Final Cut and current TVs read it fine. The BOM is a three-byte mark at the start of the file that says it is UTF-8. Some TVs, USB playback on DVD players and older Windows programs only recognize UTF-8 with that mark. If your TV keeps showing symbols, download again with a BOM.
ANSI, or Windows-1252, is only for very old devices that cannot read UTF-8 at all. It holds Western European accents but not emoji, music notes or other alphabets. For line endings, CRLF is the Windows standard and the safest for TVs; LF is the Mac and Linux standard, and current players accept both.
- Current TV or player, YouTube, video editors: UTF-8.
- Older TV that still shows symbols: UTF-8 with BOM.
- Old DVD player that rejects UTF-8: ANSI.
When an accent cannot be recovered
If the file was already saved with ? or the diamond � in place of letters, the information was gone before it reached us: the file stores the question mark, not the accent. No tool can safely guess whether caf? was café or cafe. We tell you how many words are affected so you can find the original file or fix them by hand.
To avoid this in future, always save as UTF-8. In Windows Notepad, pick UTF-8 under Encoding when saving; since 2019 it uses UTF-8 by default. When downloading subtitles, prefer files that already come as UTF-8.
More than fixing accents. Subtitles made right from your video.
Frequently asked questions
Why do my subtitles show weird characters on the TV?
Because the file was saved in one encoding, usually Windows-1252, and the TV reads it as UTF-8. Convert it to UTF-8 here; if the TV still shows symbols, download it with a BOM, which tells the TV the file is UTF-8.
How do I fix é and ñ in subtitles?
Drop the file or paste the text. Those pairs appear when UTF-8 was opened as Windows-1252 and saved again. We rebuild the letters, even if it happened more than once.
How can I tell what encoding an SRT file uses?
Drop the file and check Encoding found. We tell apart UTF-8 with and without a BOM, UTF-16, Windows-1252, ISO-8859-1 and ISO-8859-15.
What is UTF-8 with BOM?
UTF-8 with three bytes at the start that announce the encoding. The text does not change. Use it when a TV or older program still shows symbols with plain UTF-8.
Does the fix change the timing or the text?
No. Only letters with broken accents change. Numbers, timecodes and line breaks stay the same, and the list of fixed lines shows every change.
Does it work for Spanish, French or German subtitles?
Yes. Ñ, ¿, ¡, ç, ß, umlauts and every Western European accent are fixed the same way.
Want subtitles without this kind of problem? We transcribe the video.
Upload the video or paste a YouTube link. We transcribe with accents and punctuation, and you download the SRT as UTF-8, along with captioned clips.
See the result on your own recording before adding a card.