Character Encoding Inspector
Text & Unicode · Files & OS
Inspection and conversion run only in this browser. The selected file is never uploaded and never overwritten.
Drop or pick a file up to 16 MiB. The candidates appear as soon as it loads.
Read the previews and choose the candidate that reads as the intended text. Nothing is decided automatically.
Result
The result appears here after you load a file.
How to use
Load the file
Drop a text file. The BOM, a hex dump, and the reading under seven encodings appear together.
Compare the previews
Use the replacement and control counts as hints, and look for the candidate that reads naturally.
Save as UTF-8
Choose the source encoding, add a BOM if you need one, and save. A lossy conversion stops by default.
Scope of the result
Input is limited to 16 MiB, the preview to the first 2,000 characters and the hex dump to the first 256 bytes. Output is UTF-8 only; the BOM is the one choice. UTF-32 is detected but never converted. Encodings are not decided automatically, characters already replaced cannot be restored, and line endings and Unicode normalization are left alone. Support for Shift_JIS and the other legacy mappings follows the browser's own TextDecoder.
Frequently asked questions
- Does it decide the encoding for me?
- No. The same bytes can be valid in several encodings. Compare the BOM and the previews and choose.
- Does "readable without replacement" mean the encoding is right?
- No. Bytes can be valid yet still not be the intended text. Check that words and punctuation read naturally too.
- Can it repair mojibake?
- If the original bytes survive and you can pick the right encoding, yes. Anything already turned into replacement characters is gone.
- Can I save in an encoding other than UTF-8?
- Output is UTF-8 only. You can choose whether it carries a UTF-8 BOM.