Skip to main content
忆云竹

Text file encoding converter

Read legacy text files, fix encoding mismatches and export UTF-8 or UTF-16 with your choice of BOM and line endings.

Choose a text file, or drop it here.

Save file as
Line endings

Up to 10 MiB. If a legacy file looks garbled, select its original encoding. Invalid bytes stop conversion; characters are never silently replaced.

Converted text

Preview shows the first 20,000 characters. Download includes the entire file; output is limited to 40 MiB.

Processed in this browser. Nothing is uploaded or saved.

About this tool

Open an older CSV or text export that appears garbled, choose its source encoding and inspect the converted preview.

How to use

Choose a text file to preview it. The default checks for a UTF-8 or UTF-16 byte order mark; without one, it accepts only valid UTF-8. If the file is in GB18030 / GBK, Big5, Shift_JIS, EUC-JP, EUC-KR, Windows-1251 or Windows-1252, choose that source encoding yourself. A valid byte sequence alone cannot prove which legacy encoding was intended.

Choose UTF-8, UTF-16 LE or UTF-16 BE for the output. Select whether to include a BOM and whether to preserve line endings or normalize them to LF, CRLF or CR. For a CSV intended for Excel, UTF-8 with a BOM is often useful; the tool changes text encoding, not spreadsheet structure.

An incorrect source encoding can show garbled text even if decoding succeeds. Inspect the preview before downloading. Invalid byte sequences, binary control characters and conflicting BOMs stop conversion instead of replacing characters silently. Text already saved with lost characters or question marks cannot be recovered from encoding alone.

One file up to 10 MiB; output up to 40 MiB. The preview shows 20,000 characters while the download contains the complete converted file. Processing is local, cancellable and limited to 60 seconds. Files are not uploaded or stored.

Common questions

Can the original encoding always be detected?

No. Automatic mode recognizes a UTF-8/UTF-16 BOM or valid UTF-8. Select legacy encodings manually. Characters already replaced with question marks cannot be reconstructed by conversion.

Should I add a BOM or change line endings to fix garbled text?

First choose the correct source encoding and check that the preview reads correctly. A BOM identifies the output encoding; changing LF or CRLF changes line endings, not character decoding. Match those output options to the receiving application. The preview shows 20,000 characters, but downloading retains the full converted text.

Comments

0
to join the discussion.
No comments yet.
Reply thread

Follow the complete conversation chronologically; each reply identifies the specific message it answers.