Use canonical normalization first
Essential and Plain text modes apply NFC normalization, which composes canonically equivalent forms while avoiding the broader compatibility changes of NFKC.
Normalize Unicode text without guessing
Clean unusual Unicode spaces, line endings, controls, punctuation, and invisible characters with safe defaults and an exact change report.
Characters that look alike can behave differently
- A regular space, non-breaking space, narrow no-break space, figure space, thin space, and ideographic space have different code points and layout behavior.
- The same visible accented character may be stored as one code point or as a base letter followed by a combining mark.
- Line feeds, carriage returns, Unicode line separators, and paragraph separators can produce different results across systems.
- Compatibility normalization can simplify ligatures and presentation forms, but it can also change symbols that are meaningful in a specialist context.
Clean the text without losing control.
Standardize text spacing
The cleaner replaces non-breaking and unusual Unicode spaces with ordinary spaces and reports the two groups separately. Plain text can also collapse repeated spaces and replace tabs.
Remove unexpected controls
Nulls, most C0 and C1 controls, byte order marks inside text, soft hyphens, zero-width spaces, word joiners, and risky bidirectional controls are removed and counted.
Reserve compatibility cleanup
Deep clean applies NFKC and removes more format characters. This can help a strict technical destination, but it requires review when symbols, mathematical notation, emoji, or complex scripts matter.
Know what the cleanup can prove.
Stable comparison
Reduce cases where two strings look identical but fail search, equality, deduplication, or validation because their spaces or normalization forms differ.
Stable line handling
Convert mixed carriage returns and Unicode separators into consistent newline characters before a document or system interprets them.
Visible tradeoffs
See whether the cleaner changed spaces, controls, normalization, punctuation, formatting, or blank lines instead of receiving an unexplained output.
Keep formatting and provenance separate.
No general Unicode cleaner understands every domain. Source code, identifiers, cryptographic material, linguistic data, mathematical notation, and normalized database keys may require a narrower policy. Keep Essential mode for general prose and test specialist text against its real destination.
Common questions, answered plainly.
What is NFC normalization?
NFC converts canonically equivalent Unicode sequences into a consistent composed form where one exists. It generally preserves meaning better than compatibility normalization.
What is NFKC normalization?
NFKC also folds some compatibility and presentation forms. That can help strict matching, but it may change how specialist symbols or typography are represented.
Does Unicode cleanup remove accents?
No. The cleaner does not transliterate text or strip diacritics. Canonical normalization keeps the represented characters and accents.
Paste the text. See every fix.
Use the same canonical cleaner for hidden characters, unusual spaces, Markdown, HTML, smart punctuation, and clean plain-text copy.
Make copied text useful without making false promises.
Clean what is present, review what changed, and preserve any disclosure or editorial record the work still needs.
Open the AI Text Cleaner