How WPM, CPM, and accuracy are calculated.

Words per minute (WPM) is a standardized measure that counts every five typed characters as one word, regardless of the actual word lengths in the passage. The formula is straightforward: take the number of correctly typed characters, divide by five to get the equivalent word count, then divide that by the elapsed time in minutes. Two hundred and fifty correct characters in one minute equals fifty WPM. We report net WPM — the more conservative of the two common definitions — which subtracts errors from your total. Some sites report gross WPM, which counts every keystroke including wrong ones; that number is always higher and less honest about your real working speed.

Characters per minute (CPM) is the same idea without the divide-by-five normalization. CPM is more useful when comparing typing speeds across languages or scripts where word boundaries do not map cleanly to five-character chunks. In Devanagari, Tamil, and Telugu — where a single visible character may correspond to several keystrokes — CPM is often the fairer metric. Typetera displays both so you can use whichever matches the standard for your exam or workplace.

Accuracy is the percentage of correctly typed characters out of all characters you entered, including any you backspaced over and re-typed. A score of ninety-five percent means that for every twenty characters you typed, one was wrong before correction. Most professional typing benchmarks require ninety-five to ninety-eight percent accuracy as a floor — below that, speed gains stop being meaningful because you spend more time correcting than typing forward. The Results page surfaces your per-key error breakdown so you can see exactly which letters or symbols cost you the most.

Read the full WPM guide →

Three modes for three kinds of practice.

Typetera offers three distinct test modes. Each one is designed for a different use case, and switching between them takes one click — there is no setup, no account preference to save, no popup explaining which to choose.

• Sentences

  The default mode draws from a hand-curated corpus of real prose in your chosen language. Sentences are full, grammatical, and varied — not the same word repeated or random letter sequences. This is the mode to pick for honest speed and accuracy measurement, because it puts your typing under the same load you experience when writing an email, a report, or an exam answer.

• Words

  Word mode displays a continuous stream of common words from a deduplicated frequency list in your language. There is no punctuation, no capitalization, and no sentence structure — just words. This is the mode enthusiasts and competitive typists prefer because it isolates raw finger speed from the cognitive overhead of parsing meaning. It is also the mode most similar to what other typing speed sites display by default.

• Custom

  Custom mode lets you paste or upload your own passage — up to fifty kilobytes of text. This is the mode for exam prep with real past-year passages, for screening typists on the actual material they will be typing at work, or for practicing with content that means something to you. Your text never leaves the browser; we do not store it on any server, and it does not appear in your history unless you save it locally.

Six things that actually help your speed.

Practical advice, ordered roughly by how much they will move your numbers in the first month of regular practice. None of this is novel — it is the same advice serious typing teachers have given for decades.

1. 01Accuracy comes before speed, always. Most people who plateau at 40 or 60 WPM do so because they trained at the edge of their accuracy. If you are below 95 percent, slow down until you are not, and the speed will follow within a few weeks.
2. 02Look at the screen, not the keyboard. This is the single biggest jump for new typists. It feels slower for the first day or two, then permanent for the rest of your life.
3. 03Keep your fingers on the home row — left hand on ASDF, right hand on JKL semicolon. Every key has a designated finger. Returning to the home position after each reach gives you a stable launchpad and removes the visual hunting that wastes whole seconds per sentence.
4. 04Practice with real prose, not random words. Random word drills can paper over weak bigrams and rhythm. Real sentences expose where your hands actually stall.
5. 05Type every day, even five minutes. Typing speed is a motor skill, and motor skills consolidate during sleep. Five minutes a day for two weeks beats one ninety-minute session.
6. 06Stop hitting backspace for every typo. If accuracy is at 96 percent and the test rewards corrections, fine. But the habit of immediately reaching for backspace breaks rhythm and trains the wrong reflex. Practice typing forward through small errors.

Read the full speed guide →

Seven languages, native scripts.

English, Spanish, French, and Portuguese run on standard QWERTY (or AZERTY for French, where appropriate) with locale-aware punctuation and capitalization. The corpora for each language are sourced from real prose — not transliteration, not machine-translated English, not random word lists. Sentences are reviewed before they enter the test pool so that you are not typing nonsense.

Hindi (Devanagari), Tamil, and Telugu use native scripts and assume an Indic keyboard layout is active on your system — InScript, Phonetic, or a remapped layout, whichever you prefer. The first time you type in one of these scripts on Typetera, a brief layout check prompts you to confirm that the character you intend to type is the character your keyboard is producing. After that the check stays out of the way. Tamil and Telugu in particular have been weak spots on competing typing sites, and we treat them as first-class throughout.

What we do and do not collect.

Typetera does not require an account, does not ask for an email address, and does not save your typing test results to any server. Your runs are kept locally in your browser via localStorage so the Results page can show your last few attempts and your improvement over time. Clearing your browser data clears your history. Nothing is synced; nothing is uploaded.

We use Google Analytics 4 for anonymous traffic measurement so we can see which features get used and which pages are slow. Ads on supporting pages run through Google AdSense. Both can be turned off via the cookie banner at the bottom of any page; neither runs without your consent in jurisdictions that require it. The test page itself never loads ad scripts, even for users who have opted in.

Read the full privacy policy →