Typing forces eyes onto the keyboard or onto small letters on a screen. Voice removes that visual cost. StarWhisper writes into any Windows app, pairs with Magnifier, ZoomText, and high-contrast modes. Free plan, no card.
Speak the words, let Magnifier, ZoomText, or Read Aloud handle the visual layer
StarWhisper does not draw a dictation window or overlay during use. There is nothing extra on screen to find, focus, or magnify. Just a system tray icon.
One global hotkey starts and stops dictation. After the initial setup, the only visual element you need is the field you are writing into.
Magnification continues to work normally. The dictated text appears at whatever zoom level you have set, in whatever color and contrast theme is active.
Windows High Contrast themes, dark mode, custom color schemes, and screen tints all render the dictated text the same way they render typed text.
Local Mode keeps the model and the audio on your PC. No network round-trip means no spinners, no degraded mode, no waiting on a connection.
500 words per day, 3,500 per week, no credit card. Pro is $10 per month for unlimited dictation on long documents and projects.
Low vision covers a wide range of conditions: age-related macular degeneration, retinitis pigmentosa, glaucoma in later stages, diabetic retinopathy, post-stroke visual changes, recovery from eye surgery, congenital nystagmus, and many others. The shared experience is that text on a screen demands more visual effort than it demands for fully sighted users. Reading a paragraph takes longer. Spotting a typo takes longer. Finding a specific key on the keyboard takes longer, especially without large-print key labels.
Typing in particular has a hidden visual cost most people do not think about. To type efficiently, the eyes either know the keyboard so well that they never look at it (touch-typing), or they sweep between the screen and the keyboard repeatedly. For low vision users who do not touch-type, every sentence costs many of those sweeps. Each sweep is small visual fatigue. Over an hour, the fatigue adds up.
Voice input removes the sweep. The user looks only at the field they are writing into, at whatever zoom level Magnifier or ZoomText is set to. The keyboard does not need to be visible. The eyes do not need to leave the magnified region. StarWhisper sits behind a global hotkey, transcribes the speech with OpenAI Whisper, and writes the text into the focused field by simulating keystrokes. The visual surface of the interaction is just the field itself.
Windows Magnifier is the built-in screen magnifier in Windows 10 and 11. Most low vision users have it on a hotkey, with Lens mode or Docked mode tied to a familiar gesture. The interaction model with StarWhisper is straightforward: Magnifier handles the visual layer, StarWhisper handles the input layer, neither knows or cares about the other.
In practice this means a user can have Magnifier set to 300% on the active region around the cursor, focus a Word document, press the StarWhisper hotkey, dictate a paragraph, release the hotkey. The text appears in the document under the magnified region. Magnifier keeps following the cursor as new text lands. There is no setting in StarWhisper for Magnifier compatibility because nothing is needed, the two tools simply do not conflict.
For users who switch between docked and lens mode, the same pattern holds. StarWhisper does not change anything about how Magnifier sees the screen, it only inserts new text into focused fields. ZoomText, MAGic, and other third-party magnification suites behave the same way for the same reason.
Many low vision users rely on high-contrast or inverted color modes to make text legible. Windows offers system-wide High Contrast themes, individual app dark modes in Word and Outlook, and browser-level inversion or contrast extensions. None of these conflict with StarWhisper because StarWhisper writes plain text. The text inherits whatever theme the host application is using.
The StarWhisper settings window itself respects the system theme. If Windows is set to dark mode, the settings window is dark. If Windows is set to a High Contrast theme, the settings window picks up the high-contrast colors. Font sizing follows Windows display scaling, so a user running at 200% display scale sees the settings window at 200% scale. None of this is custom StarWhisper logic, it is built-in Windows behavior because the app uses standard Windows controls.
For users running custom color schemes through third-party tools like NegativeScreen or f.lux, the same logic applies. StarWhisper neither sees nor changes the color scheme. The inserted text appears under whatever color overlay or theme is active.
For users who do not touch-type, email replies are often the most visually expensive task in a daily routine. Open the reply field, look at the keyboard to find each key, look at the screen to check for typos, repeat. With StarWhisper, the routine becomes: focus the reply field at preferred zoom, press the hotkey, speak the reply, release, glance at the magnified text to check, send. The visual effort drops significantly. Most low vision users on the free plan find email volume fits comfortably within 500 words a day.
Writing a long report or document at 200% or 300% zoom means most of the visible screen is just one or two lines of text. Typing in this state is slow because the keyboard is also harder to see. Dictation removes that bottleneck. Open Word, zoom in, set the document to high contrast, dictate paragraph by paragraph. Use Microsoft Word's built-in Read Aloud feature or any third-party screen reader to read the dictated text back if needed. Pro users on $10 per month dictate unlimited words for projects that exceed the free 3,500 per week.
Online forms, especially long ones, are visually expensive. Each field demands the same sweep between keyboard and magnified region. Dictation reduces every field to a hotkey-and-speak interaction. Browsers like Chrome, Edge, Firefox, and Brave all accept StarWhisper's keystroke input the same way they accept typed input, so no browser extension is needed.
Personal writing tends to be the first thing that drops off the schedule when typing becomes too costly. With dictation, a daily journal entry in Notion or Obsidian becomes a 2-minute interaction. The free plan is comfortably enough for typical journaling volume. Users who write more move to Pro for unlimited dictation.
One question that comes up often: if I cannot easily read the screen, how do I confirm that the dictated text is correct? The answer is that any read-back tool already in your stack handles it. Windows Narrator can read the inserted text on demand. Microsoft Word and Edge both have built-in Read Aloud features that read any text in the document or page. NVDA, JAWS, and ZoomText Reader all read inserted text the same way they read typed text. StarWhisper does not need to add a read-back feature because the host application's tools cover it.
For users who prefer to mostly use voice for input and visual review for output at high magnification, the pattern is: dictate, zoom in on the inserted paragraph, scan for obvious errors, correct with the keyboard, continue. For users who prefer audio-only review, the pattern is: dictate, hit the Read Aloud key in Word or Edge, listen, correct with the keyboard or by re-dictating, continue. Both patterns are common, both work.
StarWhisper's accuracy with OpenAI Whisper is high enough that most paragraphs need few corrections. For more on the accuracy comparison with other dictation tools, see the accuracy page and the Whisper local vs cloud FAQ.
A common misconception is that voice input solves all eye fatigue. It does not. Voice removes the visual cost of producing text, but the eyes still do the work of reviewing the text afterward, and the eyes still do the work of every other visual task in the day. Voice input is one tool in a larger set of eye-strain reduction strategies, alongside larger fonts, better contrast, regular zoom-out breaks, proper monitor distance, the 20-20-20 rule, and rest.
That said, even a partial reduction in visual cost compounds. A user who dictates email instead of typing it saves a measurable amount of eye time per day. Over weeks, the saved fatigue is often noticeable, which is part of why dictation tends to stick once a low vision user finds one that works without fighting their existing setup. StarWhisper's design goal is exactly that: be the part of the stack that lowers visual cost, not the part that adds to it. For overlapping use cases see the carpal tunnel page, the blind users page, and the dyslexia page.
Other ways voice input lowers the friction of writing on Windows
Works alongside JAWS, NVDA, and Narrator. Hotkey-driven, no overlay, screen reader reads the result.
Close the gap between thought speed and typing speed. Free, local, in any Windows app.
Skip the spelling tax. Dictate, screen reader or eye reads back, edit lightly.
One hotkey writes to any Windows text field. No per-app plugin, no per-app permission.