Unicode Normalization Tool

Convert text to NFC, NFD, NFKC or NFKD and inspect whether its exact character sequence changed.

Text processing stays in this browser. Input is limited to 20,000 UTF-16 code units.

NFC is the default. NFKC and NFKD remove compatibility distinctions and can change meaningful formatting. No case folding or accent stripping.

Normalize equivalent text sequences

Text can look the same while using different Unicode sequences. An accented letter may be stored as one precomposed code point or as a base letter plus a combining mark. This tool applies one of the four Unicode normalization forms and reports whether the exact sequence changed. It is useful before comparing text under a deliberately chosen normalization policy.

Choose the form your workflow requires

Paste text into Input text, choose Normalization form, then select Process text. The default is NFC. Load example (replaces input and settings) restores NFC and inserts a decomposed accent together with compatibility characters.

Form Operation Example consequence
NFC Canonical decomposition followed by composition e + U+0301 becomes é
NFD Canonical decomposition é becomes e + U+0301
NFKC Compatibility decomposition followed by composition ① becomes 1; A becomes A
NFKD Compatibility decomposition Compatibility changes plus decomposed accents

For the loaded Café ① A, NFC produces Café ① A; the output contains eight code points. NFKC instead produces Café 1 A. The tool does not strip accents, translate letters, change case or sort text. An unchanged result means the exact input sequence already matches the selected normalization output, not that it has no Unicode complexity.

Preserve distinctions intentionally

Compatibility forms can remove distinctions that matter to typography, identifiers or source material. Keep an original copy before applying NFKC or NFKD to stored data. Normalization alone does not resolve look-alike letters from different scripts and is not a security filter. Its use in account names, search keys or data matching needs a policy appropriate to that system.

The normalization behavior follows the browser's JavaScript implementation. Unicode's normalization FAQ explains canonical versus compatibility equivalence, and MDN documents the four normalize forms. These distinctions explain why a compatibility conversion should not be treated as a harmless visual cleanup.

Review and save the result

Copy full result copies normalized text, and Download result saves UTF-8 TXT. The result preserves spaces and received line breaks; it adds no final newline. Browser textareas may turn pasted CRLF or CR into LF before normalization. Input is limited to 20,000 UTF-16 code units; normalization can expand the output. Unpaired surrogates produce an error because they cannot be faithfully exported as Unicode scalar text.

If two strings still differ after the same normalization form, inspect their individual values with the Unicode Code Point Inspector. For visible-character grouping rather than equivalence, use the Grapheme Cluster Viewer. Rerun after changing the form: the previous output is cleared to prevent copying a result from an earlier setting.