Split Multilingual Text Into Words
Admin ·
A word list based on spaces alone can leave a whole Thai phrase as one item. Choose the source language so your browser can find likely word boundaries, then decide whether a downstream search or analysis should include punctuation and emoji.
Find the boundary inside ภาษาไทย
Choose Load example. The sample begins ภาษาไทย! and also contains Café with a combining accent, 0012 and a scientist emoji. The example selects Thai. Choose One segment per line, leave Include punctuation and symbols off, and select Split words.
The result has four lines, in order ภาษา, ไทย, Café, and 0012. The first two words are adjacent in the source. The accent and the leading zeros stay intact.
ECMAScript Internationalization API, Intl.SegmenterUnicode text segmentation and word boundaries
Keep punctuation and emoji in the JSON export
Enable Include punctuation and symbols and select JSON array. Split again. The exclamation mark appears after ไทย and the scientist emoji becomes the last segment. The export contains six strings, including the original 0012.
Open the download with a JSON parser to check the array order and types. Choose Space-separated text for a plain string or One segment per line for manual review.
ECMAScript Internationalization API, Intl.SegmenterUnicode text segmentation and word boundaries
Verify the source order of Arabic words
Paste مراجعة، جاهز 42 and select Arabic. With punctuation excluded, the JSON array is ordered مراجعة, جاهز, and 42. Enable punctuation to add the Arabic comma after the first word. The position in the array follows the stored source order even when text appears right to left on screen.
The browser supplies the language segmentation. Dictionary boundaries and treatment of compounds can vary by browser version. Compare unusual names or multilingual compounds with your intended analysis.
ECMAScript Internationalization API, Intl.SegmenterUnicode text segmentation and word boundariesUnicode bidirectional text display
Common questions
Will the tool change the spelling of the exported words?
It exports the substrings from your input without changing case, expanding contractions, or normalizing characters.
Does the language selector translate the text?
It selects the browser’s word-boundary rules. The original language and characters are preserved.