Word Counter

Word counts matter more than most writers expect. School and university assignments specify a minimum or maximum word count; job applications cap cover letters at a certain length; SEO meta descriptions and titles have hard character limits before search engines truncate them with an ellipsis; social media posts have platform-enforced character ceilings; and scripts, presentations and podcasts are often budgeted in minutes, which only makes sense once you know how many words that time actually holds. A live counter that updates as you type removes the guesswork and the constant trips to a word processor's status bar. This tool reports seven figures at once. Character count with spaces is the rawest measure — every character you typed, including spaces, tabs and line breaks, counted using JavaScript's Unicode-aware iteration (`Array.from`) rather than raw string length, so accented letters and most emoji count as one character each instead of being miscounted when they are represented internally by two UTF-16 code units. Character count without spaces strips whitespace before counting, which is the number most character-limit fields (like a database column or a tweet, historically) actually care about. Word count splits the text on runs of whitespace and discards empty fragments, so multiple spaces between words never inflate the count and leading/trailing whitespace is ignored. Sentence count looks for `.`, `!` or `?` followed by whitespace or the end of the text, treating each such boundary as the end of a sentence, then discards any resulting empty fragment (which happens, for instance, if the text ends with punctuation and nothing after it). This is a heuristic, not true natural-language sentence segmentation — it does not know that "Dr. Smith" or "e.g." contain a period that is not a sentence break, so texts with many abbreviations will show a slightly inflated sentence count. For the vast majority of ordinary prose, though, this simple rule tracks the real sentence count closely. Paragraph count splits the text wherever there is a blank line (one or more line breaks with only whitespace between them), then discards empty paragraphs, so a document with extra blank lines between paragraphs is not over-counted. A single block of text with no blank lines at all counts as one paragraph, matching how most word processors would report it too. Reading time and speaking time are both derived from the word count using standard, widely cited average rates: 200 words per minute for silent reading, and 130 words per minute for speaking (a natural presentation or podcast pace, slower than reading because speech includes pauses and emphasis that reading does not). Both are shown as a whole number of minutes, rounded up, with a `< 1 min` display when the estimate comes out to less than a full minute — for a 40-word paragraph, telling you "1 min read" would overstate a text that actually takes about 12 seconds to read, so the tool is explicit about sub-minute content instead of always rounding up to a whole minute. Both estimates are averages, not measurements of your specific reading or speaking speed — technical or dense text reads more slowly than casual prose, and a practiced public speaker delivers faster than someone reading unfamiliar material aloud for the first time. Treat the numbers as a useful ballpark for planning (should this blog post be split into two? Will this speech fit in a five-minute slot?) rather than a stopwatch-accurate prediction.