Paste articles, drafts, or exports and see which words repeat. Tokenization is Unicode-aware (letters, numbers, apostrophes). Filter with English stop words, a minimum length, and optional case folding. View a sortable table with counts and percentages, or Copy TSV for spreadsheets. All processing stays in your browser — nothing is uploaded to DroidXP.
Enter text to see word counts.
| Word | Count | % of counted |
|---|
It breaks your input into word tokens and counts how many times each distinct token appears. That is useful for editing repetitive prose, spotting overused terms, comparing drafts, or getting a quick lexical summary before deeper NLP tooling. It is not a substitute for professional corpus linguistics or search-engine ranking models.
Tokens are runs of Unicode letters, marks (accents), numbers, and apostrophes; punctuation and spaces split words (so hyphenated words usually become two tokens). Optional stop-word removal drops a built-in list of common English function words. Minimum length ignores very short tokens. Fold to lowercase merges different capitalizations of the same spelling.
Count is how many times that token appeared after filters. % of counted is that count divided by the total number of token occurrences after filters — not raw document length. Use
Copy TSV to paste a full word, count, percent table into Excel or Google Sheets (the on-page table may list only the first 400 rows for speed).
Pair with the Character Counter for length limits, Text Sorter for line order, and the Duplicate Line Remover when you need unique lines before counting.
No. Counting runs entirely in your browser. Drafts may be saved in localStorage on this device only.
Contiguous runs of Unicode letters, combining marks, numbers, and apostrophes. Punctuation and whitespace separate tokens. Hyphens usually split tokens because they are not included in the word pattern.
Each percent is the share of counted token occurrences: count divided by the total occurrences after your filters, not the raw character length of the document.
A small built-in English list of common function words (articles, pronouns, conjunctions, and similar). It is a convenience filter for prose, not a linguistics-grade stop list. Turn it off when you need every token.
By default words are folded to lowercase so Hello and hello merge. Uncheck “Fold to lowercase” to keep distinct capitalizations as separate entries.
The on-page table shows the first 400 rows for performance. Copy TSV always exports the full sorted list for your current filters.
Keys are prefixed with droidxp-wfc-: input text, case option, stop words, minimum length, and sort mode. Clear removes them.
Unicode letters in many scripts are tokenized when your browser supports Unicode property escapes in regular expressions. Stop-word filtering is English-only; disable it for other languages.
It is related: you see how often terms appear. Search engines use far more than raw repetition; use this as a writing or editing aid, not a ranking guarantee.
Follow your policy. Nothing is sent to DroidXP, but screen capture, extensions, and backups can still expose clipboard or page content.
After the page is cached, it can work offline. No network is required for counting.