Reading Time Calculator estimates how long your text takes to read, listen to as an audiobook, deliver as a speech, or skim — all in your browser and updated as you type. It uses the peer-reviewed silent-reading baselines from Trauzettel-Klosinski & Dietz (2012, Investigative Ophthalmology & Visual Science) for 15 languages, adjusted to match Brysbaert's (2019) meta-analysis of 190 English studies (238 wpm). A content-type multiplier accounts for the difficulty of what you're reading — technical docs at 0.65× baseline, poetry at 0.45×, blog posts at 1.0×. A reader-comparison table shows how times change for slow, average, fast, and speed readers. Everything runs client-side and works offline — nothing is uploaded.
What this tool measures
The calculator returns four core times for any pasted text:
- Reading time — silent reading, adjusted for language and content type.
- Listening time — how long a professional audiobook narrator would take at 160 wpm (Audible baseline), with a 1.5× playback figure for people who speed up their audiobooks.
- Speaking time — a presentation pace of 150 wpm (roughly the natural cadence of an unhurried keynote).
- Skim time — 450 wpm, the pace of someone scanning for the gist rather than reading every word.
It also shows word, character, sentence, and paragraph counts, plus a comparison table with five distinct reader profiles so you can see how much your audience matters.
The reading speed baselines
Silent reading speed for English is set to 238 words per minute, the pooled estimate from Marc Brysbaert's 2019 meta-analysis of 190 studies covering over 17,000 readers (Journal of Memory and Language). Reading aloud runs slower at 183 wpm — most audiobook narrators sit near that figure with performance factored in.
For other languages, the baselines come from Trauzettel-Klosinski & Dietz (2012), which measured silent reading rates for a standardised text (IReST) in 17 languages. Chinese and Japanese sit around 158–193 wpm; Romance languages cluster near 180–220 wpm; Arabic and Finnish are slower due to script and morphology.
These are population averages. A fluent, college-educated reader on familiar prose is usually 20–30% faster; a non-native or fatigued reader is 20–40% slower. Use the reader comparison table to bracket the estimate.
Why content type changes the number
Words per minute is not a property of the reader alone — it's a property of the reader-and-text pair. The same person reads a blog post at 240 wpm and a mathematics paper at 90 wpm without noticing anything unusual. The tool applies a difficulty multiplier to the baseline so estimates stay realistic:
- Blog / news: 1.0–1.05× — familiar structure, everyday vocabulary.
- Fiction: 1.0× — narrative flow, few backtracks.
- Non-fiction book: 0.85× — more to absorb, slower pace.
- Academic paper: 0.75× — dense argument, footnotes, citations.
- Technical / docs: 0.65× — code blocks, jargon, frequent scroll-backs.
- Poetry: 0.45× — line breaks, re-reads, deliberate pauses.
These multipliers are calibrated from published reading-rate studies and the WPM ranges Nielsen Norman Group publishes for web content.
Why include a reading time on your blog post
Medium popularised the "5 min read" badge because their own experiments found it improved article completion rates. The intuition matches what UX researchers have measured elsewhere: a visible time estimate lets a reader decide up front to commit — which increases scroll depth, reduces early bounces, and slightly improves time-on-page. All three are signals Google's ranking systems care about, if indirectly.
Practical rules of thumb: keep the label above the fold, use plain minutes (not seconds or fractional minutes), and pair it with word count only if your audience skews technical. A copy-ready line at the bottom of the tool gives you an easy paste-and-go format.
Reading time vs speaking time vs listening time
These three are commonly confused but represent very different tasks:
- Silent reading (238 wpm) is fast because the eye skips over familiar words and small grammatical particles without registering them individually.
- Speaking a script (130–160 wpm) is slower because articulation is rate-limited by the mouth. News anchors hit ~150 wpm; conversational speech averages 130. Auctioneers and rap vocalists reach 250+ wpm but at the cost of comprehension.
- Audiobook narration (~160 wpm) is deliberately slower than natural speech so listeners can absorb prose without a paper text to anchor them. The tool includes a 1.5× playback figure because a substantial share of audiobook listeners speed up their audio.
If you're timing a keynote, use the speaking figure and add 10–15% for pauses, laughs, and audience questions. If you're producing a podcast, the speaking figure is a floor — natural conversation runs longer than a read script.
Privacy and offline use
The calculator runs entirely in your browser. Your text is never sent to a server, never logged, never stored. You can disconnect from the internet, paste an unpublished draft or a confidential document, and the times will still update as you type. Open the Network tab in your browser's dev tools and confirm — zero requests fire when you type.