For RSI recovery, accessibility, or pure speed: a complete hands-free workflow on Windows. Free voice dictation app, optional foot pedal, 130 words per minute without touching the keyboard.
Free software, optional hardware, full coverage of writing tasks.
Download StarWhisper from the homepage. Free to install, no signup. The first launch downloads the Whisper model (around 200 MB). After that the app stays in the system tray and waits for the hotkey. During first-run setup, pick a microphone (a headset or USB mic gives better accuracy than a laptop's built-in mic) and choose a hotkey. For hands-free use you will replace that hotkey with something foot or voice-triggered in the next step.
Three real options depending on budget and preference.
Option A, foot pedal (best for sustained use): A USB or Bluetooth foot pedal mapped to the same key you set as the StarWhisper hotkey. Recommended hardware: the Stream Deck Pedal (~$90, three pedals, fully programmable via Elgato software), the Kinesis Savant Elite 2 (~$100, ergonomic, popular among RSI users), or a budget single-pedal USB unit from Olympus or other vendors (25 to 40 dollars). Map one pedal to your StarWhisper hotkey. Press and hold the pedal to dictate, release to stop.
Option B, Windows Sticky Keys (free): Enable Sticky Keys in Windows Accessibility settings. Tap your StarWhisper hotkey once to "stick" it down (no need to hold), tap again to release. Less smooth than a pedal but completely free.
Option C, Windows Voice Access (free, fully voice-controlled): Pair StarWhisper with the built-in Voice Access (Win+H or Settings, Accessibility, Voice Access). Voice Access provides the click/scroll/navigation layer; StarWhisper provides the higher-accuracy dictation layer when you say "start dictating".
Open Word, Outlook, Chrome, Slack, Notion, your code editor, or any text field on Windows. Click into the text field (or use your hands-free clicking option from step 2). Press the foot pedal (or trigger your chosen activator). Speak. Release. The transcription is typed in like keyboard input. Works in every Windows app. Many users find a rhythm of pedal-down, one sentence, pedal-up, glance at the screen, pedal-down for the next sentence.
For light edits, re-dictate the sentence (Whisper usually gets it on the second pass when you speak more clearly). For navigation, Voice Access supports arrow keys, click commands, and a grid overlay for precise clicking. A foot pedal can be mapped to common shortcuts (Ctrl+Z, Ctrl+S, Tab) for one-step editing. For long-form drafts, many writers accept the rough first draft and revise on a later pass either with help or in shorter editing sessions. The point is to keep typing-related strain to zero during dictation.
Real workflows that work without keyboard input.
Cutting keyboard load by 70 to 90 percent during recovery is often the difference between healing and surgery. Many ergonomists and physical therapists recommend dictation as part of the standard RSI protocol. See the voice-to-text for carpal tunnel guide for more.
For users with motor impairments, missing limbs, or paralysis, hands-free typing is permanent infrastructure rather than a recovery tool. StarWhisper plus Windows Voice Access plus eye tracking can cover the full input stack.
Sustained dictation at 130 to 150 WPM is roughly 3x average typing speed. For people who write long-form (authors, bloggers, support agents, journalists) the speedup compounds. The first draft of a 1500-word post can land in 15 to 20 minutes.
Walking pads, treadmill desks, and standing desks all work poorly with keyboards (typing while walking is awkward). Dictation removes the keyboard from the loop, making walking-while-working actually viable for hours.
Hands-free coverage extends to 96 languages with auto-detection. Useful for multilingual writers, translators, and international knowledge workers. Language support details.
Audio is processed locally on CPU or GPU. Nothing uploaded. Important for medical, legal, financial, or otherwise confidential content. Privacy architecture.
For two decades, hands-free typing was theoretically possible but practically frustrating. Dragon NaturallySpeaking was the gold standard, but it cost 700 dollars and required hours of voice training to reach acceptable accuracy. Windows Speech Recognition was free but accuracy was poor, especially with accented speech. The result: most people who needed hands-free input ended up either using Dragon under duress or living with reduced output.
OpenAI Whisper changed that. Released in 2022 as an open-source model, Whisper hit 95 to 98 percent accuracy out of the box across most accents and conditions, with zero per-user voice training. Run locally on a modern CPU or any NVIDIA GPU, it transcribes faster than people speak. For the first time, a serious hands-free workflow can be assembled from free software plus optional consumer hardware (a foot pedal, a decent USB mic) and deliver Dragon-tier accuracy without the Dragon price tag or training overhead.
This guide covers the full setup. StarWhisper provides the dictation layer. A foot pedal (or Sticky Keys, or Voice Access) provides the trigger. Optional accessibility tools handle mouse work. The result is a workstation that needs no typing at all for the vast majority of knowledge work.
For sustained hands-free work, a hardware trigger you can operate with your foot is the single biggest comfort upgrade. Approximate prices and tradeoffs:
| Hardware | Price | Best For | Tradeoff |
|---|---|---|---|
| Stream Deck Pedal | ~$90 | Three pedals, fully programmable, mainstream support | Wired USB, slightly bulky |
| Kinesis Savant Elite 2 | ~$100 | Ergonomic shape, popular in RSI community | Wired USB, premium price |
| Olympus RS27/RS28 | ~$50 | Transcriptionist-grade, durable | Originally for transcription, fewer modern reviews |
| Generic USB foot pedal | $25-40 | Lowest budget, mappable in software | Build quality varies, may need extra config |
| Bluetooth pedal (various) | $60-120 | Cable-free, works with laptops | Battery to manage, occasional pairing issues |
| Sticky Keys (no hardware) | $0 | Zero budget, works immediately | Two-tap pattern less smooth than pedal |
For someone testing the workflow before committing, start with Sticky Keys (free, immediate). If you settle into hands-free work and want it to be permanent, the Stream Deck Pedal is the most-supported mainstream choice. RSI-recovery users tend to prefer the Kinesis Savant Elite 2 for the ergonomic pedal shape.
Microphone choice matters more than you might expect. A USB headset mic (Logitech H390, Audio-Technica ATR2100x, or any decent gaming headset) gives much better accuracy than a laptop's built-in mic because it stays close to your mouth and rejects room noise. Whisper handles imperfect audio well, but a 100 dollar mic eliminates almost all transcription errors and pays for itself in saved correction time within a few weeks of sustained use.
For someone in active recovery from repetitive strain injury, carpal tunnel, or tendinitis, the goal is to drop keyboard load to near zero while staying productive enough to keep working. A practical protocol:
Most people see a 70 to 90 percent drop in keystrokes per day with this protocol. That is usually enough for the inflammation cycle to break, allowing physical recovery while preserving income. For deeper context on the medical side, see the carpal tunnel typing alternative reference page.
Sustained dictation produces real productivity wins, but the size of the win depends heavily on what you write and how much editing it needs. Honest estimates from real users:
The dominant gain is not speed; it is endurance. People who dictate often can sustain 6 to 8 hours of writing per day without physical fatigue, where typing the same volume would be exhausting and probably injurious. For full-time writers, journalists, and customer-support agents, this matters more than peak speed.
For people who want to test the full hands-free workflow without buying any hardware, the free Windows stack is genuinely usable.
Built into Windows 11 (and downloadable for Windows 10). Open Settings, Accessibility, Voice Access, and turn it on. Provides voice commands for clicking, scrolling, switching apps, and basic dictation. Use it for the click layer of your workflow. The dictation built into Voice Access is okay but uses Microsoft's older speech engine, which is why pairing it with StarWhisper makes sense: Voice Access handles clicks, StarWhisper handles the words.
Open Settings, Accessibility, Keyboard, Sticky Keys, turn on. Now a hotkey press "sticks" until you press it again. Combined with the StarWhisper hotkey, you tap once to start dictating and tap once to stop. Without a foot pedal or pedal-substitute, this is the lowest-friction way to make StarWhisper hands-free.
The free stack: Sticky Keys to control StarWhisper's dictation trigger, plus Voice Access for clicking and navigation. Zero hardware cost. The whole workflow runs on the average Windows 11 laptop. Hardware upgrades (foot pedal, USB mic) improve comfort and accuracy but are not required to start. For the comparison with Windows Voice Access alone, see the StarWhisper vs Windows Voice Typing page.
Honest disclosure. Even a strong hands-free workflow does not cover everything. Activities where typing or precise mouse work is still the practical answer:
For most knowledge workers, the percentage of time spent in these categories is small. For those who do a lot of design, video editing, or hardcore coding, the realistic outcome is a hybrid setup: hands-free for the 80 percent of time spent writing, reading, and replying, with hands available for the 20 percent of time spent in precision-critical tools. That hybrid is often enough to break an RSI cycle.
Detailed RSI-recovery workflow with hardware and pacing tips.
Problem-aware reference for evaluating dictation against other options.
Universal text-field coverage across Office, browsers, IDEs, chat.
Accuracy and feature comparison with the built-in Windows option.