Subtitle files showing mojibake like "测试" or question marks? Your SRT probably has the wrong encoding. SRTKit detects GBK, Big5, UTF-8 and Latin-1 mismatches and rewrites your file cleanly. No sign-up. No forced watermark.
Fix My SRT File — FreeWorks on .srt, .vtt and .ass · Processes in your browser · [TODO:实测→定稿 retention policy]
You exported an SRT on a Windows machine and now every Chinese line reads as random symbols. Classic GBK-vs-UTF-8 mismatch — easily reversible if you know the source encoding.
A wall of "???" or empty squares means the editor couldn't read the bytes at all. Usually a Latin-1 file mislabeled as UTF-8. We detect and re-decode in place.
Traditional-Chinese subtitles saved as Big5 often look scrambled on modern players. Convert Big5 to UTF-8 without losing a single character.
French or Spanish accents appearing as "é" or "ñ"? That's UTF-8 bytes being read as Latin-1. One-click re-encoding back to readable text.
SRTKit probes your file's byte patterns to guess the real encoding — UTF-8, GBK, Big5, Latin-1 or Windows-1252. No need to know in advance what went wrong.
Modern players, video platforms and browsers all expect UTF-8. We transcode your file to clean UTF-8 while preserving every cue index, timestamp and line break.
Only the character bytes change — your cue numbers, timecodes and formatting stay byte-identical. Drop the result straight back into your editor.
Fixing dozens of episode files? Use the batch tool to re-encode an entire folder at once. Same detection logic, same clean output.
The repaired file is yours, unmodified beyond encoding. No injected headers, no banner cues, no forced branding — just a working SRT.
Most SRT encoding issues come from a mismatch between the file's actual byte encoding and what the player expects. A file saved as GBK on Windows and then opened as UTF-8 on macOS — or vice versa — produces mojibake. Re-saving without specifying the source encoding makes it worse.
Upload your GBK-encoded .srt file to SRTKit. The tool detects the GBK byte signature, decodes the text correctly, and re-encodes everything as UTF-8. The output is readable in every modern subtitle editor and video player.
Yes. Encoding repair only touches the human-readable text bytes. Cue indices, timecodes, formatting tags and line breaks remain untouched.
Yes. The same detection-and-re-encode logic works on WebVTT (.vtt), SubStation Alpha (.ass/.ssa) and most other text-based subtitle formats. For subtitle formats with complex styling, run the file through the converter first to normalize.
[TODO:实测→定稿 single-file and batch limits]. For larger libraries, the batch tool handles multiple files in one go.
Drop your file in, get a clean UTF-8 SRT out. Free, no sign-up, no forced watermark.
Start the Encoding Fix