Sentence Counter
Paste or type your text below, sentence count and average words per sentence update instantly, with every detected sentence listed below.
About the Sentence Counter
Sentences are split on ., !, or ? followed by whitespace or the end of the text, so you can see at a glance how many you've written and exactly how each one got detected.
- Sentences, split on
.,!or?followed by a space/line break or the end of the text. - Avg words / sentence, total word count divided by sentence count, a common readability signal.
- This is a simple free-tool heuristic and doesn't special-case abbreviations like "Dr." or "e.g.", for most everyday text it works well.
A decimal number, price, or version number mid-sentence can cause text to vanish, not just miscount
The detection pattern requires a period, exclamation point, or question mark to be immediately followed by whitespace or the end of the text before it's treated as a sentence boundary — a period sitting directly in front of another digit, like the one in "3.14" or "$4.99," fails that check and is correctly not read as an ending. But because the pattern has no other way to consume text across an embedded period like that, the consequence isn't just "no split there" — it's that the words leading up to and including that embedded period can be silently dropped from the results entirely. Running "The price is $4.99 for one." through this tool produces a single listed sentence reading "99 for one.," with "The price is $4." missing from the output altogether, not merged into the sentence before or after it. The same thing happens with version numbers ("Version 2.0 is out now." loses "Version 2." from the list) and with any other mid-sentence decimal point.
Abbreviations still get treated as their own short sentence
The whitespace-after-punctuation check that catches embedded decimals doesn't help with abbreviations, since "Dr." followed by a space and a capitalized name looks structurally identical to a real sentence ending. "Dr. Smith arrived early." gets split into two listed entries, "Dr." and "Smith arrived early.," the same kind of false split that essentially every lightweight, non-dictionary-based sentence detector produces on abbreviations, initials, and honorifics.
How this differs from this site's Readability Checker
The Readability Checker uses a simpler pattern with no requirement that a terminator be followed by whitespace, which means it never drops text the way this tool can — but it also means it treats every embedded decimal point as a full sentence break, inflating its sentence count on numeric text more than this tool does. This tool's whitespace lookahead avoids that specific over-counting problem, at the cost of occasionally losing a fragment of text around an embedded decimal instead. Neither approach is a full grammatical sentence detector; they just fail in different, specific ways.
The average words-per-sentence figure can read misleadingly high on affected text
The word count used for the average comes from splitting your full original text on whitespace, independent of how the sentence-splitting regex behaved — so if a passage contains several sentences with embedded decimals that get dropped or merged, the sentence count in the denominator can end up lower than the text's actual number of sentences, while the word count in the numerator stays accurate for the whole passage. That combination pushes the reported average words-per-sentence upward, past what the text would actually measure if every sentence had been detected correctly.
The numbered list exists so you can spot detection problems, not just read a total
Rather than showing only a sentence count, the tool lists out every detected sentence individually in a numbered list built with real DOM elements. That's deliberate: a bare number gives you no way to notice when something like an abbreviation or an embedded decimal has caused a false split or dropped text, but scanning the actual list makes those problems immediately visible — an unusually short entry like "Dr." or a sentence that clearly starts mid-thought is a visible sign the detection heuristic tripped up on that particular spot, which a single summary statistic would hide completely.
Frequently Asked Questions
Why is part of my sentence missing from the list, like "The price is $4." disappearing entirely?
This happens with sentences containing an embedded decimal point, like a price or version number. The tool correctly avoids treating that period as a sentence ending, but its pattern has no other way to include the text leading up to it, so that portion gets dropped from the results rather than merged into a neighboring sentence.
Does an abbreviation like "Dr." get counted as its own sentence?
Yes. Since "Dr." is followed by a space and a capitalized word, it looks structurally identical to a genuine sentence ending, so it gets listed as a short standalone sentence. This is a common limitation of lightweight, punctuation-based sentence detectors.
How is this different from the Readability Checker's sentence detection?
The Readability Checker's pattern has no whitespace requirement after punctuation, so it never drops text the way this tool can, but it does treat every embedded decimal point as a sentence break, inflating its count on numeric text. This tool avoids that over-counting at the cost of occasionally losing a fragment around a decimal point instead.
Can the average words-per-sentence number be misleading?
Yes, on text with embedded decimals. The word count comes from your full original text, but the sentence count can undercount actual sentences when some get dropped or merged around a decimal point, which pushes the reported average higher than the text would actually measure with every sentence correctly detected.
Is my text sent to a server when I use this tool?
No. All detection and counting happens locally in your browser as you type; nothing is uploaded anywhere.