Fix text and CSV encoding
When accents and special characters come out as strings of nonsense, the file was read with the wrong encoding. This tool converts files saved in old Windows encodings to UTF-8 and repairs double-encoded text, with automatic detection for the common cases. Optionally adds the marker Excel needs to read accents correctly. It runs on your device.
Drop files here or click to browse
Results
How to fix text and CSV encoding
- Drop the CSV or text file with the broken characters.
- Leave the source encoding on automatic, or pick Windows-1252, ISO-8859-1 or repair double-encoding.
- Choose whether to add a BOM for Excel.
- Run and download the corrected UTF-8 file.
Why accents turn into nonsense
If a file shows two or three strange symbols where an accented letter, an apostrophe or a currency sign should be, it has been read with the wrong character encoding. Text is stored as bytes, and an encoding is the rule for turning bytes into letters. Modern systems use UTF-8. Older Windows programs, and Excel in particular, often save and load text using a regional encoding such as Windows-1252. When a file written one way is read the other way, every character outside plain ASCII comes out wrong. The data is not lost; it is just being decoded with the wrong rule.
Two different problems, two fixes
The file is in a legacy encoding. A CSV exported from an old Windows tool is stored as Windows-1252 or ISO-8859-1. Tools that expect UTF-8 show question marks or replacement characters. Fix: tell this tool the source encoding and it re-saves the file as UTF-8, which everything reads correctly.
The file is double-encoded, or "mojibake". Someone read a UTF-8 file as Windows-1252 and saved it again, so the correct bytes were re-encoded into two or three wrong characters each. This is where the classic capital A with a tilde before every accent comes from. Fix: the repair option reverses the mistake by re-encoding those characters back to bytes and decoding them properly as UTF-8.
Automatic detection
With the source set to automatic, the tool first tries to read the file as strict UTF-8. If that fails, it falls back to Windows-1252. If it succeeds but the text contains the tell-tale patterns of double encoding, it applies the repair. In most cases this is all you need; set the encoding by hand only if the automatic result still looks wrong.
The BOM option
Excel on Windows still assumes a regional encoding when opening a CSV, unless the file starts with a byte order mark. Switch the BOM on when the file is going back into Excel; leave it off for databases, programming tools and most web imports, some of which treat the marker as part of the first column name.
What you get
| Symptom | After repair |
|---|---|
| Names with accents show a stray capital A plus a symbol in place of each accented letter | Müller, José, Zoë |
| Apostrophes and quotes appear as three odd characters | It's, "quoted" |
| The euro sign appears as a small cluster of symbols | € 25 |
| Question marks or boxes where letters should be | The original letters, once the right source encoding is chosen |
The output is a UTF-8 text file with the same name and content, only correctly encoded. It works for CSV, TXT and any other plain text file. Nothing is uploaded; the repair runs in your browser.
Frequently asked questions
Why do accented letters show as strange symbols?
The file was written as UTF-8 and read as a Windows encoding, or the reverse. The characters are not lost, just decoded with the wrong rule, and this tool reverses it.
Is the file uploaded?
No. The repair happens in your browser and the file stays on your device.
Will it work for languages other than European ones?
Yes for anything that was originally UTF-8 and got double-encoded. For files genuinely saved in a regional encoding, the tool covers the Western European ones.
Should I add the BOM?
Yes if the file is going back into Excel on Windows, no for databases and programming tools.
Your files never leave your device
Every tool on Squishly Tools is 100% free and runs entirely inside your browser. Your images and documents are processed on your own computer and are never uploaded to a server. There are no accounts, no watermarks and no catch.