How Multilingual Word and Character Counting Works
Accurate text statistics require more than splitting on whitespace. While English, Spanish, and French rely on spaces between words, East Asian languages such as Chinese (Simplified & Traditional) and Japanese do not use spaces to delimit words. Our tool uses native browser Intl.Segmenter lexical boundary analysis backed by a deterministic morpheme fallback to guarantee exact counts across all scripts.
Standard Text Metrics Explained
- Words: Segmented word-like tokens according to locale rules (e.g. CJK morphemes, Latin words, hyphenated compounds).
- Characters (with spaces): Total perceived Unicode grapheme clusters, ensuring emojis and multi-byte code points are counted truthfully as single visual characters.
- Characters (without spaces): Pure content characters excluding spaces, tabs, line breaks, and full-width ideological spaces.
- Sentences & Paragraphs: Identified using locale-aware terminal punctuation (such as full stops, question marks, and CJK
。！？) and non-empty line break blocks.
Scientific Reading and Speaking Time Estimates
Estimates are calculated using documented empirical speeds: ~200–250 words per minute (wpm) for Latin reading, ~130–150 wpm for speaking, and ~300–500 characters per minute (cpm) for CJK text. Durations are rounded to realistic, intuitive intervals without implying false second-level precision (such as "< 1 min" or "2 min").
Frequently Asked Questions
- Le compteur prend-il en charge les textes chinois et japonais ?
- Oui. Grâce à la segmentation Unicode (Intl.Segmenter), les textes CJK sont analysés précisément sans dépendre des espaces.
- Comment le temps de lecture est-il calculé ?
- À partir de moyennes éprouvées (~225 mots/min en texte latin) avec des arrondis sans précision de seconde trompeuse.
- Mon texte est-il transféré vers un serveur distant ?
- Non, tous les calculs sont effectués à 100 % en local dans votre navigateur.