The tool decodes one byte sample as UTF-8, Big5, GB18030, Shift_JIS and Windows-1252 when the browser supports them. Names, punctuation and column boundaries are usually more useful than an automatic guess.
A matching BOM is strong evidence. Replacement characters, controls and suspicious sequences reveal some bad decodings, but several encodings can still produce readable-looking text. You make the final choice.
Only the selected encoding is applied to the complete file. CSV and TSV downloads include a UTF-8 BOM for spreadsheet compatibility; TXT downloads use plain UTF-8. No text is uploaded.
Common questions and answers about this topic.
Add the original CSV, compare the candidate previews, check familiar names and columns, then choose the readable encoding and download the UTF-8 copy. Keep the source until you verify the result.
Use a matching BOM when present, then inspect actual content. The best candidate should preserve expected names, punctuation and separators with few or no replacement characters. The order alone is not proof.
No. XLSX is a structured binary workbook, not a plain text file. Export the sheet as CSV first, or use spreadsheet software that can choose the source encoding during import.
The BOM helps spreadsheet applications recognize UTF-8 when a CSV or TSV is opened directly. TXT output stays plain UTF-8 because text editors generally detect it without that marker.
No. Reading, previewing and UTF-8 conversion run in this browser. The source is not uploaded or overwritten. Closing or reloading the page clears the working state.
No. The tool can reinterpret existing bytes and save the chosen text as UTF-8. It cannot reconstruct characters that were replaced, deleted or exported incorrectly before this file was created.