To calculate reading time of an article, divide the total number of words by an appropriate words-per-minute rate; the answer is the same arithmetic whether you are timing a silent reader, an audience member following along, or a presenter reading a script aloud. The formula is simple, but the two questions most people ask are not the same question: a text that takes one adult four minutes to read silently can easily take five or six minutes to deliver aloud, because speech is slower than silent reading for almost everyone. The default silent-reading rate used by research-grade tools comes from a 2019 meta-analysis by Marc Brysbaert, published in the Journal of Memory and Language, which aggregated 190 studies and arrived at 238 words per minute for average adult silent reading of English non-fiction, with fiction slightly faster at 260 wpm. Reading aloud is slower still: the same meta-analysis derived 183 wpm from 77 oral-reading studies. Picking the right rate for the right task, and seeing the word count that produced the number, is what turns a guess into a defensible reading-time estimate. That is exactly what the Reading Time Calculator is built to do.
Silent Reading vs Speaking Aloud: Two Different Questions
Most reading-time widgets collapse silent reading and spoken delivery into a single number, but they are different tasks that produce different speeds. Silent reading lets the eyes skip ahead, skim familiar phrases, and backtrack without losing your place, and the median adult reader processes English prose at roughly 238 words per minute when measured across hundreds of studies. Reading aloud removes most of those shortcuts. The mouth has to articulate every word, the speaker has to breathe, and listeners need time to follow unfamiliar names and numbers, which is why oral-reading rates cluster around 183 wpm in the same body of research.
When you calculate reading time of an article for a blog post, a landing page, or a "5 min read" badge, the silent figure is the one your audience will experience. When you calculate reading time of an article for a podcast script, a wedding toast, a class reading, or a stage presentation, the spoken figure is the one that determines whether you fit inside your slot. Using the wrong rate for the job is the most common reason reading-time estimates feel "wrong" in practice, and showing both side by side is the simplest way to avoid that mistake.
How to Use the Reading Time Calculator
- Paste your text into the input box, or switch to word-count mode and type the number of words directly if the article lives in another file or you already know the count.
- Pick a preset reading rate for the silent estimate, or type your own words-per-minute value, then set the speaking rate for the read-aloud estimate.
- Read both durations in the result panel and check the word count and counting rule shown underneath, so every number in the result is auditable.
The calculator shows silent and spoken durations in the same view, each at its own rate, which lets you decide which figure matters for the situation you are planning. Because the word count is displayed alongside the rule the calculator used, you can copy the text into another tool, count it the same way, and arrive at the same answer, or notice precisely where the two tools disagree.
Where the Words-Per-Minute Numbers Come From
The presets shipped with the calculator are not folklore. They are pinned to published measurements so the defaults have a defensible source rather than a guess:
| Use case | Words per minute | Source |
|---|---|---|
| Silent reading, English non-fiction (average adult) | 238 wpm | Brysbaert 2019, meta-analysis of 190 studies |
| Silent reading, English fiction (average adult) | 260 wpm | Brysbaert 2019, meta-analysis of 190 studies |
| Oral reading, average adult | 183 wpm | Brysbaert 2019, derived from 77 oral-reading studies |
| Standardized reading aloud (IReST texts) | 228 wpm | Trauzettel-Klosinski & Dietz 2012, IReST study (IOVS) |
| Presentation / speech convention | 150 wpm | Speech-coach convention; no retrievable study asserts this figure |
Two of these numbers are easy to confuse. The 183 wpm oral-reading figure is what Brysbaert's review reports when people read naturally aloud; the 228 wpm IReST figure comes from a very different protocol in which participants read short standardized paragraphs under a stopwatch. Because the tasks are different, both are offered rather than collapsed. The familiar 150 wpm "presentation" rate is also present, but it is labelled honestly as a coaching convention rather than a measured average, because no published study actually asserts it.
A quick worked example makes the arithmetic visible. Take a 1,000-word article. Divided by the 238 wpm silent-reading preset, 1,000 ÷ 238 ≈ 4.20 minutes, which the calculator rounds to 4 min 12 sec. At the 183 wpm oral-reading rate the same article comes out to 1,000 ÷ 183 ≈ 5.46 minutes, or 5 min 28 sec. The two numbers are both correct; they answer two different questions about the same text.
If none of the presets fit your situation, ignore them and type any rate of your own. A second-language learner reading dense technical English at 120 wpm, or a fast scanner skimming marketing copy at 450 wpm, both deserve a reading-time estimate built on the rate that actually applies to them rather than a population average.
Why Reading Time Tools Disagree With Each Other
Reading-time estimates are nothing more than words divided by a rate. When two tools arrive at noticeably different answers for the same article, the disagreement almost always lives in the word count rather than the division, because every calculator agrees on the math but not on what counts as a word. Hyphenation, contractions, decimal numbers, thousands separators, and stray punctuation each behave differently across tools, and the gaps can add up to several percent on a long piece.
The Reading Time Calculator follows a stated rule rather than a vague notion of a word: text is split on whitespace after word-joining dashes are treated as separators, and a token counts as a word when it contains at least one letter or digit. Under that rule, "state-of-the-art" counts as one word, "don't" counts as one, "3.14" counts as one, "1,000" counts as one, and a string of punctuation alone counts as none. The same rule is printed under the result, so if a number surprises you, you can see exactly why it came out that way.
That visibility is the point. A reading-time estimate you can audit by hand is one you can defend in an editorial meeting, a client review, or a planning spreadsheet. A reading-time estimate that hides its inputs is only useful until someone asks how it was calculated, and at that point you either have to trust the tool or recompute from scratch.
Common Use Cases for a Reading Time Estimate
Calculating the reading time of an article is useful in more situations than the typical "5 min read" badge at the top of a blog post. A few that come up regularly:
- Editorial planning. A target of 1,500 words at the 238 wpm silent rate works out to about six minutes, which is a common upper bound for a Medium-style long read. Comparing that figure against your draft tells you whether to cut, or whether the target is the wrong target.
- Presentation timing. A 20-minute conference slot at the 150 wpm convention is roughly 3,000 words; at the 183 wpm oral-reading average it is about 3,660. The gap between those two numbers is the gap between a talk that finishes early and one that overruns.
- Script budgeting. Podcast intros, video voiceovers, and audiobooks are all priced and timed against the spoken rate. If you have ever had to trim a 14-minute script into a 10-minute slot, the only honest way to do it is on a per-word basis against the rate the host actually speaks at.
- Reading lists and assignments. Students and instructors benefit from the same arithmetic. A 4,000-word reading that the syllabus labels "30 minutes" is closer to 17 minutes at 238 wpm silent, which can change how a week's workload actually adds up.
In every one of these cases, the figure that matters is the figure produced by the rate that matches the situation, which is why a tool that exposes both silent and spoken durations is more useful than one that picks a single rate and hides it.
How to Read the Result Honestly
The numbers the calculator returns are population averages, not promises. Brysbaert's own review puts most adults between 175 and 300 words per minute silently, which means the true reading time for any particular reader can differ from the estimate by roughly a third in either direction. A four-minute estimate is honest for an average reader; it is six minutes for a slow reader and under three for a fast one. There is no rate you can pick that fixes that spread, because it reflects real variation in vocabulary, topic familiarity, fatigue, screen size, and interest.
That limitation is not unique to this calculator; it is a property of every reading-time estimate that is honest about being one. The calculator is precise about its arithmetic, open about its inputs, and explicit about which preset is a measured average and which is a coaching convention. Use the result to compare drafts against each other, to budget a presentation slot, and to set expectations for readers, not to promise anyone an exact number of minutes.
Everything runs in the browser; the text is never uploaded, stored, or attached to an account. That matters when you are timing a draft you have not yet published, a confidential client document, or a script you would rather not hand to a server. The estimate you see on screen is computed locally, and the only thing that leaves your machine is whatever you choose to do with the number.