Average typing speed is about 40 words per minute. Average conversational speech is about 150. If you stop typing first drafts and start speaking them, you nearly 4x your raw output without learning a new skill. StarWhisper makes this work in any Windows app.
It is not your brain. It is the keyboard sitting between your brain and the page.
Average adult typing speed sits at roughly 40 words per minute. Even a fast professional typist tops out around 80. Meanwhile, your brain composes sentences at conversational speed, somewhere between 120 and 180 WPM. The gap is real and it shows up in two ways.
StarWhisper turns any Windows app into a dictation surface. Hold a hotkey, talk, release, and the text appears where your cursor is. First drafts come out at near-conversational speed. You still edit by typing, where typing is actually good.
The math is simple, the workflow is two changes, and the apps you already use still work.
You speak 3 to 4 times faster than you type. That is not a marketing number, it is a measured one. Average typing is 40 WPM. Average conversational speech is 150. Even after editing time, net output is well above pure typing for most writers.
Typing forces you to look at each word as it appears, which triggers premature editing. Dictation skips that loop. You speak the rough sentence, it appears, you keep going. The mess is fixable later. The momentum is not recoverable once you lose it.
The first paragraph is the hardest because the cost of typing 200 words you will throw away feels high. Spoken, that 200 words takes 80 seconds. Throwing away 80 seconds of breath is not the same psychological cost as throwing away 5 minutes of typing.
Writers, copywriters, journalists, and students all develop wrist strain after years of high-volume typing. Voice cuts your daily keystroke count in half overnight. Many users start dictating specifically because of RSI and stay for the speed.
StarWhisper uses the Windows IME and auto-paste, so it works in any application that accepts text input. Notion, Word, Google Docs, Substack, Scrivener, your email client, your CRM, your chat tool, your IDE, every browser tab.
OpenAI Whisper, the engine under StarWhisper, was trained on roughly 680,000 hours of multilingual audio. Word accuracy on clear English audio is in the 95 to 98 percent range. You will not spend more time fixing transcription errors than you saved by speaking.
Typing speed has been measured for decades, and the consensus numbers are remarkably stable. The average adult types at about 40 words per minute on a standard QWERTY keyboard. A trained professional typist sits around 65 to 75. A very fast typist hits 100. The world record on Monkeytype is over 300, but that is a one-minute sprint, not sustainable output. Sustained writing speed for almost everyone is far below the burst speed they could hit on a typing test.
Speech rates have been measured almost as long. Average conversational English sits around 150 words per minute. Newscasters speak at about 140 to 160 because that is the rate that feels natural to listen to. Auctioneers reach 250 to 350 for brief bursts. Comfortable spontaneous speech for almost any adult sits in the 120 to 180 range. That is two to four times your sustained typing speed.
If you take your typical 90-minute writing session and replace the first-draft typing with dictation, you do not get a 1.5x improvement, you get something closer to 3x. A 1,500-word draft that took 90 minutes to type takes 30 to 40 minutes to dictate. Add 15 to 20 minutes of editing afterward. Total time: about 50 minutes for the same output, with hands that hurt less and a brain that is less drained from staring at a screen.
The "real writers type their own sentences" myth is recent and false. Some of the most productive writers in history outsourced the physical writing step entirely, either to a secretary or to a recording device.
Schwartz is one of the most-cited direct-response copywriters in marketing history. His book Breakthrough Advertising is still considered the canonical text on long-form sales copy. His working method was famously fast and famously dictation-heavy. He used a kitchen timer set to short bursts, dictated copy aloud, and considered the typing-it-up step a separate, slower task. His sales letters generated tens of millions in revenue for clients in the 1960s and 1970s.
Asimov published over 500 books across science fiction, popular science, and non-fiction. He was famously prolific and openly described the working method behind it. Early in his career he typed everything himself. Later, as the output volume grew, he dictated significant portions to a secretary or directly into a recorder. His point in interviews was always that the bottleneck is composition, not finger speed, and that a writer who insists on typing their own first drafts is choosing a slower workflow for aesthetic reasons rather than practical ones.
Churchill dictated almost all of his major political writing and history books to secretaries. Barbara Cartland wrote over 700 romance novels, almost all of them dictated. Henry James dictated his late-period novels to a typist after handwriting started giving him cramps. The list goes on. The pattern is that high-output writers eventually find that the keyboard, or the pen, is the slowest part of their pipeline and they move past it.
The mistake most first-time dictation users make is trying to dictate a finished sentence. That is not what dictation is for. Dictation is for getting a rough first draft out of your head and onto the page at conversational speed. Editing is still a typing task, because editing rewards precision and small movements and that is what keyboards are good at.
Spend 5 minutes typing a bullet outline of what you want to say. This is the part of writing where you actually think hardest. Voice does not help here because the limiting factor is your structure, not your finger speed.
Open the target app. Place your cursor where the section starts. Hold the StarWhisper hotkey. Speak the section out loud, end to end, without trying to polish. Release the hotkey. The text appears. If you mess up mid-sentence, just say "no, scratch that" or restart the sentence; you will clean it up in the edit pass. Move to the next section.
Walk away from the screen. The break matters because editing is much faster when you are not still mentally inside the draft.
Come back. Read through the rough dictated draft. Cut the false starts. Tighten the prose. Restructure the bits that came out backwards. This is where keyboards are actually fast, because edits are small targeted movements, not sustained throughput.
Net result for a 1,500-word piece: 5 minutes outline, 12 minutes dictation, 5 minutes break, 25 minutes editing. Total 47 minutes for a draft that would have taken 90 to 120 minutes of pure typing.
Not all writing benefits equally from voice. Here is the rough hierarchy.
| Writing type | Speed gain from voice | Notes |
|---|---|---|
| Long-form blog posts and articles | 3x to 4x | Conversational tone, low formatting |
| Email and Slack/Teams messages | 3x to 5x | Where most writers spend the most time |
| Marketing copy and sales letters | 3x to 4x | Spoken sounds more natural anyway |
| Fiction and creative prose | 2x to 3x | Voice helps with dialogue rhythm |
| Student essays and research papers | 2x to 3x | Citations still typed manually |
| Note-taking and journaling | 3x to 4x | Especially effective for daily notes |
| Technical documentation | 1.5x to 2x | Code blocks still typed |
| Code itself | 0x to 1x | Type code, dictate comments and commit messages |
| Legal contracts, citations | 0x to 1x | Precision punctuation, type these |
Voice is great for prose and bad for precision. If you are typing code, every bracket and colon matters and saying "open paren close paren semicolon" is slower than just typing it. If you are drafting a legal contract where the exact punctuation has legal weight, you want the deliberate pace of typing. If you are doing heavy formatting work in a structured document with lots of bullet levels and table cells, voice loses to keyboard shortcuts. Editing in general is faster by keyboard than by voice. The dictate-first edit-later workflow is built around this reality: voice for high-throughput first drafts, typing for everything that needs surgical precision.
For everything else, which is the bulk of what most writers, copywriters, students, and knowledge workers actually produce, voice wins on time.
Total time to first dictated paragraph: about 5 minutes.
If you searched "how to write faster" or "write faster software," you probably fit into one of these patterns.
In every one of those cases, voice for first drafts is a near-zero-risk experiment. The download is free, the trial is a single 5-minute install, and the speed difference is large enough that you will know inside one session whether it works for you.
If you want to write faster and you only do one thing, switch to voice for first drafts. The math is too lopsided to argue with. 40 WPM typing versus 150 WPM speaking is a 3.75x raw output difference, and even after editing time the net stays well above pure typing. Famous high-output writers across every era figured this out and worked around the keyboard. Modern speech engines like Whisper made the same workflow available to everyone without a personal secretary.
StarWhisper is free for personal use, $10 per month for unlimited Pro, runs locally on Windows 10 and 11, and works in every app you already write in. It is the cheapest possible test of whether voice writing fits your brain. Most writers who try it for a week stop typing first drafts entirely.
Real numbers, real workflows, and the edit-time math.
If your bottleneck is meetings, not articles, start here.
Workflow guide for novelists, bloggers, and substack writers.
Why high-output copywriters dictate first drafts and edit second.